the effect of brand-consumer congruency on brand adoption...
TRANSCRIPT
1
THE EFFECT OF BRAND-CONSUMER CONGRUENCY ON BRAND ADOPTION BEHAVIOR IN SOCIAL NETWORKING SITES
By
STEFANIE RIEDIGER
A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ADVERTISING
UNIVERSITY OF FLORIDA
2010
2
© 2010 Stefanie Riediger
3
ACKNOWLEDGMENTS
I would like to express a heartfelt thank you and much gratitude to everyone who
has helped me in achieving my Master of Advertising degree. First and foremost, I am
greatly in debt to the extended invaluable and diligent hands-on assistance of my thesis
chair, Dr. John Sutherland. He agreed to take on advisor duties and pick up where
another advisor left off when it was not required of him. His unyielding patience and
encouragement was monumental. His guidance pushed me to produce quality work and
helped me believe in my abilities to do so. I would also like express my gratitude to Dr.
Jorge Villegas, Dr. Jon Morris and Dr. Robyn Goodman for their time and guidance with
regard to my study. Further thanks go to Jody Hedge for her administrative help.
I would also like to thank my loving parents who have been supportive of me every
step of the way, pushing me to do my best. Their continued love and support helped
remind me of what was important when the going got tough and discouragement set in.
I could always count on them to keep me focused on my goals and what needed to be
done to achieve them. I would also like to extend my thanks to my sister, Allison, my
brother-in-law, Kevin, and many wonderful friends who were sources of encouragement
in this process. They were always there for me.
Lastly, I would like to thank those from my past who helped me realize my
capabilities as a student, a professional, and most importantly as a person. They made
me see that I really can work through anything by taking things one step at a time.
4
TABLE OF CONTENTS page
ACKNOWLEDGMENTS .................................................................................................. 3
LIST OF TABLES ............................................................................................................ 6
LIST OF FIGURES .......................................................................................................... 7
ABSTRACT ..................................................................................................................... 8
CHAPTER
1 INTRODUCTION ...................................................................................................... 9
2 LITERATURE REVIEW .......................................................................................... 13
A Digital Revolution ................................................................................................ 13 Participatory Cultures.............................................................................................. 13 Uses and Gratifications ........................................................................................... 18 Online Profiling ....................................................................................................... 20 The New Word-of-Mouth ......................................................................................... 23 Self-Congruity Theory ............................................................................................. 25 Hypotheses ............................................................................................................. 28
3 METHOD ................................................................................................................ 34
Research Design .................................................................................................... 34 Participants ............................................................................................................. 35 Data Collection and Procedure ............................................................................... 36 Measures and Instrument ....................................................................................... 40
Independent Variable ....................................................................................... 42 Dependent Variable .......................................................................................... 45 Motivation Interaction Variables ....................................................................... 47
4 RESULTS ............................................................................................................... 48
Data Analysis .......................................................................................................... 48 Sample Profile .................................................................................................. 48 Facebook Tendencies/Activity .......................................................................... 49 Brand Attitudes, Congruity, and Compatibility .................................................. 50 Independent, Dependent, and Interaction Variables ........................................ 51 Dependent Variable: Adoptive Behavior ........................................................... 52
Hypothesis Testing ................................................................................................. 53 Interaction Effect of SNS Motivations ..................................................................... 57 Research Questions ............................................................................................... 61 Limitations ............................................................................................................... 66
5
5 DISCUSSION AND CONCLUSIONS ...................................................................... 82
Discussion .............................................................................................................. 82 Implications ............................................................................................................. 83 Implications Summary............................................................................................. 97 Future Research ..................................................................................................... 99
APPENDIX
A INVITATION .......................................................................................................... 103
B ONLINE SURVEY QUESTIONNAIRE .................................................................. 104
LIST OF REFERENCES ............................................................................................. 110
BIOGRAPHICAL SKETCH .......................................................................................... 115
6
LIST OF TABLES
Table page 4-1 Sample profile summary statistics ...................................................................... 68
4-2 Current Facebook profile behavior (direct expression) ....................................... 70
4-3 Current Facebook profile behavior (indirect expression) .................................... 71
4-4 Brand attitudes, brand congruity, and compatibility ............................................ 72
4-5 Summary statistics for independent and motivation variables ............................ 73
4-6 Dependent variable component correlations ...................................................... 73
4-7 Dependent variable measures ............................................................................ 74
4-8 Correlations ........................................................................................................ 75
4-9 Self-brand congruity correlations ........................................................................ 75
4-10 Compatibility correlations ................................................................................... 76
4-11 Multiple regression for entertainment motivation variable (H3) ........................... 77
4-12 Multiple regression for information motivation variable (H4) ............................... 78
4-13 Multiple regression for social interaction motivation variable (RQ1) ................... 79
4-14 Multiple regression for convenience motivation variable (RQ2) .......................... 80
4-15 Summary of multiple regression significant relationships ................................... 81
7
LIST OF FIGURES
Figure page 2-1 Overview of the study ......................................................................................... 33
8
Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Advertising
THE EFFECT OF BRAND-CONSUMER CONGRUENCY
ON BRAND ADOPTION BEHAVIOR IN SOCIAL NETWORKING SITES
By
Stefanie Riediger
August 2010
Chair: John Sutherland Major: Advertising
The emergence of social networking sites as a medium for communication has
provided a wave of word-of-mouth opportunities for advertisers. This study investigates
the self-congruity theory and social networking site (SNS) usage motivations as
contributing factors of branded posting behavior on SNSs through regression analysis.
This research revealed that brand attitudes, self-brand congruity, and motivational
effects are unique per brand, making context and relevancy of brands important factors
for marketing campaigns on SNSs. Compatibility was not significantly related to
branded SNS posting behavior, but a significant positive relationship occurred between
self-brand congruity and the likelihood of adoptive posting behavior. All four motivations
tested also had significant relationships with adoption behavior likelihood acting
independently, but none of the motivations interacted significantly with self-brand
congruity in predicting behavior. Thus, to increase likelihood of SNS word-of-mouth,
individuals must increasingly feel as though the typical user of a brand mirrors their own
self-image. Independent of self-brand congruity, SNS usages for entertainment,
information, social interaction, and convenience purposes can also indicate increased
adoption behavior, depending on the brand in question.
9
CHAPTER 1 INTRODUCTION
To further develop the most successful brand communication strategies, it is
helpful to study the connections made by a consumer that can ultimately lead to
branded behavior and purchase decisions. This study is designed to observe social
networking site (SNS) posting behavior and the intricacies of both consumer
perceptions and attitudes toward a brand to explain what factors contribute to the
likelihood of branded adoption behavior on SNS. This objective was achieved by
examining the effects of self-brand congruity and brand-SNS compatibility on branded
posting behavior within an SNS.
Previous studies have examined congruity in terms of a person’s actual self-image
in comparison with his/her ideal image (Parker, 2005; Sirgy et al., 1997). Parker (2005)
also applied this comparison to brand personalities. The most relevant portion of this
research, however, involves the integration of these ideas by looking at the alignment
between self-image and brand-user image. Measures of congruity between one’s self-
image and a brand-user image are determined as a result of this integration (Parker,
2005; Sirgy et al., 1997). The resulting self-brand congruity influences consumer
behavior as a function of the brand attitude elements of familiarity, attachment, and
loyalty (Parker, 2005). The self-congruity model within the realm of consumer behavior
will act as a pivotal variable, along with the perceived compatibility of a brand with a
particular SNS, as possible explanations for SNS posting behavior.
To facilitate the explanation of why consumers interact with brands within SNSs,
the uses and gratifications theory was adapted to the model. The assumption involved
in this theory is that media users are active users who take actions in search of a goal
10
(Blumler & Katz, 1974). Furthermore, Ko, Cho, and Roberts (2005) argued that these
people “are aware of their needs, and select the appropriate media to gratify their
needs” (p. 60). The following investigation of the translation of consumer identification
with a brand into adoption behavior of a brand relies heavily on formulated insight based
on this theory’s foundational assumptions. The assumption to be explored is that the
higher a person’s self-brand congruity (or the more a consumer identifies his/her actual
and ideal self with the typical user of a brand), the more likely that person is to adopt the
brand into his/her SNS posting behavior. In addition, the uses and gratifications theory
is applied with the prediction that a person’s motivation for using an SNS could affect
how that person perceives brands within that context and ultimately how that person
posts.
Thus, the final element of this study, which contributes a new perspective to the
research on consumer attitudes and behaviors, is the medium within which consumers
are observed. The Internet has a growing and encouraging capacity for interaction that
has intensified the uses and gratifications theory (Ko, Cho, & Roberts, 2005). Not only
has the Internet been a natural incubator for many types of relationships and interaction,
but more specifically, SNSs have emerged as prominent stimulators of social activity
(Boyd, 2006). Until the development of online communities, connectivity and interaction
was limited to one-way or two-way communication (Shirky, 2009). However, with the
development of SNSs in recent years, this capacity has multiplied exponentially due to
the power of mass speech given to the individual. Webs of one-way and many-way
communication are formed simultaneously, creating an entire new world in which
11
marketers can immerse themselves and capitalize on an opportunity for premium
word-of-mouth communication value (Shirky, 2009).
One outcome of the social networking age is the practice of online profiling. The
analytical nature of the Internet makes a succinct, yet encompassing first impression a
commanding deliverable. The resulting trend of profiling oneself on Internet
communities continues to grow, as SNSs such as Facebook facilitate a diverse range of
communications (Willett, 2009). As these types of sites become increasingly popular,
new advantages of reaching consumers through this medium present themselves to
marketers. The self-branding that millions of SNS members engage in invites an
opportunity to research a brand’s self-brand congruity role in the process. This study
aims to explain self-brand congruity’s role by answering some unexamined fundamental
questions involving the consumer’s core drive for branded posting behavior on SNSs.
Many studies have been executed regarding usage of the Internet, but a relatively
limited number of these have been specified to SNSs. Even fewer studies have related
the self-congruity theory to SNS behavior. The following research expands upon
knowledge regarding posting behavior on SNSs, particularly Facebook. These cognitive
connections are important for marketers because their past habits are becoming
obsolete as strategies transition from mostly push to mostly pull by the consumer. It is
also important to gain a better understanding of the thought process behind brand
adoptive posting behavior because these actions carry word-of-mouth trust that is
greatly valued in today’s marketing practices.
The following research collected data from undergraduate students at the
University of Florida and registered users of the SNS, Facebook. The data were
12
collected by an online survey method using www.surveymonkey.com. The
questionnaire involved was designed to measure Facebook familiarity and attitudes,
brand familiarity and attitudes, self-brand congruity, brand-Facebook compatibility, and
motivations for using Facebook as possible influential factors of branded adoption
behavior likelihood. Facebook was chosen as the SNS of interest in the hope of
projecting key findings onto SNSs as a whole. The two brands observed in this study
were Starbucks and Dunkin’ Donuts. The resulting data were analyzed with the help of
the computer software formerly known as Predictive Analytics Software (PASW
Statistics 18 or PASW), and now known as IBM SPSS Statistics (SPSS). Correlations
and relationships between variables were measured using regression analysis to test
hypotheses and explore research questions that emerged from the review of the
literature of consumer behavior, social psychology, brand management, and new media
communications.
13
CHAPTER 2 LITERATURE REVIEW
The introduction of social networking sites (SNSs) to the Internet has brought with
it unchartered territory. For those with mass communication interests, SNSs have
introduced questions and possibilities in areas such as consumer behavior, social
psychology, brand management, and new media communications. The present study
aims to provide answers to a few of these questions by looking principally at motivations
and self-congruity as possible explanations for SNS behaviors.
A Digital Revolution
Social Networking sites are creating a new revolution within the broader Internet
revolution that has taken place during the last 20 years (Shirky, 2009). The Internet has
conveniently provided the platform for social networks to develop. In the following
excerpt, Shirky (2009) discussed the social networking implications of the Internet from
a filmed presentation titled How Social Media Can Make History:
The Internet is the first medium in history that has native support for groups and conversation at the same time. Whereas the phone gave us the one-to-one pattern, and the television, radio, magazines, books gave us the one-to-many pattern. The Internet gave us the many-to-many pattern. . . . As all media gets digitized, the Internet also becomes the mode of carriage for all other media. . . . That means that every medium is right next door to every other medium. Internet. . .is increasingly more a site of coordination. (3:47)
Participatory Cultures
While increased coordination has certainly facilitated globalization and changed
cognitive connections, people are primarily concerned with the cultures in which they
participate (Boyd, 2006). Thus, the focus has evolved on participatory cultures. The
Convergence Culture Consortium (2006) defines participatory culture as follows:
Participatory culture describes the way consumers interact with media content, media producers and each other as they explore the resources
14
available to them in the expanded media landscape. Consumers become active participants in shaping the creation, circulation and interpretation of media content. Such experiences deepen the consumer’s emotional investment in the media property, and expand their awareness of both content and brand. (p. n.p.)
This increased awareness and active participation involves uploading content, adapting
existing content, working collaboratively with others, and forming online connections as
core activities (Willett, 2009). As participants in this respect, Internet users are
considered to be “active agents” who display competency, personality, and knowledge
(Beavis, 2009).
Social networking sites play an important role in this agency and participatory
development, taking the Internet’s coordination function mentioned by Shirky (2009) a
step further. An SNS is a type of online community and participatory culture which
typically follows a basic format consisting of a personal profile, a blog and/or message
board, a private messaging utility, and a news feed that reports friends’ recent activities
(Boyd & Ellison, 2007; Li, 2008). Boyd and Ellison (2007) identified three defining
features of SNSs. They allow individuals to 1) construct a profile that is at least semi-
public, 2) formulate a list of other users with which they share a connection, and 3)
navigate and interact with other users’ profiles within that list (Boyd & Ellison, 2007).
Many different SNSs are available these days for a variety of purposes. The
nature and nomenclature of connections via SNSs may vary from site to site, but they
all generally serve the basic function of facilitating connections (Boyd & Ellison, 2007).
For example, LinkedIn, founded in 2002 (Company History, 2008), is an SNS with the
specific purpose of making business and career connections. Some SNSs serve to
unite those with certain health conditions such as Juvination, which brings Type 1
diabetics together to share stories, tips, and information. Furthermore, some sites such
15
as Friendster, which originally aimed to inspire more romantic connections, were
transformed by entertainers as a platform for their promotions (Boyd & Ellison, 2007).
Friendster ultimately ran into its demise by not embracing this new direction, but
MySpace was able to successfully capitalize on the networking opportunity in its wake
(Boyd & Ellison, 2007). Moreover, Facebook was initially intended to serve a niche in
the Harvard University-only community before its popularity transitioned it into a global
phenomenon.
Since their launches, MySpace and Facebook have emerged as the leading
communities with the highest number of members (Lopez, 2008). As of 2006, MySpace
totaled 100 million users and has continued to linger around 100 million unique users in
the period since (Arrington, 2009). Likewise, MySpace’s highest rival, Facebook,
grossed 18 million users in 2007 with “an estimated 90% of all undergraduates at
colleges and universities where Facebook is available are registered users of the site.
Of those users, 60% log on daily” (Lopez, 2008, p. 25). In June 2009, Facebook tallied
more than 200 million active users with more than 100 million of those logging on at
least once per day (Statistics, 2009). Most recently, less than one year later, Facebook
reported more than 350 million active users in December 2009 with the average user
spending 55 minutes of his/her day on the site (Statistics, 2009).
Facebook, initially launched in February 2004, is an SNS that aims to help people
connect with others and the world around them (About Facebook, 2004). As with many
other SNSs, Facebook allows users to share information and media such as photos,
videos, and links to other online media. By providing these functions, Facebook allows
users to develop and maintain relationships with friends and family on a consistent
16
basis (About Facebook, 2004). Typically, these personalized networks are mostly
comprised of connections previously made offline, meaning that SNSs, Facebook in
particular, are less frequently used to make new connections (Boyd & Ellison, 2007).
Facebook is unique from other SNSs, however, in that it originally began as an SNS
solely for Harvard University students before expanding to other universities. A person
was required to have a verifiable college email address in order to register with the site.
Facebook later expanded its registration requirement to all valid email addresses by
September 2006 (Company Timeline, 2009).
According to the third party audience analysis site, www.quantcast.com (2010),
facebook.com traffic is skewed slightly female (55%) and heavily Caucasian (75%), with
the highest trafficking age group being 18 to 34 year olds (42%). Furthermore, when
compared with similar SNSs, Facebook has the highest percentage of traffic to have a
college education, accounting for 53% of Facebook traffic (facebook.com - Quantcast
Audience Profile, 2010). The initial exclusivity to college students, the foundation of
college-educated users, and its top-ranked popularity among SNSs are all reasons
Facebook was chosen for this study. For a sample of primarily undergraduate college
students and recent graduates, Facebook is most likely to present respondents already
familiar with the SNS.
With such social utilities as the ones provided by Facebook at hand, SNSs can be
used for a variety of purposes. According to Boyd (2008) on the subject of political
action and SNS behavior, our society is “status-obsessed and narcissistic” and “typical
social network site users are more invested in adding glitter to pages and SuperPoking
their ‘friends’ than engaging in any form of civically driven collective action” (p. 241).
17
Research has indicated that approximately 40% of SNS users choose to include a
self-description, invitation to contact, hobbies and interests, and references to friends
and relationships in their online profiles (Van Cleemput, 2009). Posting photographs is
another highly employed self-identifying behavior observed. Lange and Lampe (2008)
discovered that 88.9% of the participants they observed displayed an identifying
personal photograph. With so much personal information available in one location,
another study found, not surprisingly, that “participants use Facebook to learn more
about others, as the level of disclosure on Facebook may be higher and have more
detail than disclosure in real life” (Lange & Lampe, 2008, p. 15).
Interchangeability between online and offline communication and activity is also a
common theme in SNS posting behavior. For example, participants tend to use
Facebook to coordinate offline, real-life activities (Lange & Lampe, 2008). Transposable
offline worlds for online worlds are corroborated by the fact that most SNS users draw
on SNSs “to gather with friends when physical co-presences is impossible or
impractical. . . . For many of the most active participants on social network sites,
networked publics substitute for physical publics because physical publics are
inaccessible, untenable, heavily regulated, or downright oppressive” (Boyd, 2008,
p. 242). These are a few of the SNS posting trends observed thus far. However, many
questions have been left unanswered as new trends add and transform existing ones. A
basic understanding from pre-behavior explanations to post behavior observations
would help provide detailed answers to these questions. Thus, in looking at a
foundational measure, motivations and the uses and gratifications perspective can help
attach meaning to these purposes for SNS behavior.
18
Uses and Gratifications
Research has indicated that motivations can be of great consequence to SNS
behavior (Beavis, 2009). The uses and gratifications theory states that varying reasons
exist to use a mass medium (Blumler & Katz, 1974). This theory explores intrinsic needs
and motivations that lead to an individual’s choices to perform specific behaviors
(Blumler & Katz, 1974). The behaviors chosen serve to satisfy those needs; needs
which can vary greatly between Internet usage and for other mediums (Ko, Cho, &
Roberts, 2005). This theory assumes that media users are active users who take
actions in search of a goal (Blumler & Katz, 1974). Furthermore, Ko, Cho, and Roberts
(2005) argued that these people “are aware of their needs, and select the appropriate
media to gratify their needs” (p. 60). For example, Wu (2009) stated that regarding
political behavior on Facebook, “endorsement of either an entertainment celebrity or a
political figure indeed is first and foremost built upon strong affective affiliation, hence
rendering a strong urge in the supporters to vocally cheer their heroes” (p. 17). This
urge to express support for their heroes represents the need in this case, and since the
features of Facebook provide extended opportunity to share political ideas and other
original political materials such as videos and transcripts of speeches (Wu, 2009),
Facebook became the appropriate medium used to gratify this need.
Thus, the Internet is a growing and encouraging capacity for interaction that, as a
result, has intensified the uses and gratifications theory (Ko, Cho, & Roberts, 2005).
Korgaonkar and Wolin (1999) looked at 41 items related to Internet usage and concerns
and found indications that consumers use the Internet for retrieving not only information,
but also entertainment and escape. Furthermore, a previous study looked at the
influence of Internet usage motivations on adoption behavior toward online services.
19
This study showed, depending on the type of website (information, infortainment,
shopping), that different motivators were key influencers on behavior (Lin, 1999).
Applying this same logic to SNSs would indicate that different users of SNSs could
view SNSs in different capacities depending on their motivation for utility. Just as with
various websites, SNSs can serve many functions within themselves. Therefore, it
should follow that the same concept of matching different motivational influencers to
varying levels or types of behavior may be applicable within the more specific context of
SNSs. As a result, some motivations may have a greater interaction effect on SNS
behavior than other motivations.
In concert with this, the impact of motivations is further amplified by the
interactivity of the Internet. The Internet allows users to participate at an elevated level
by giving them control, according to their preferences and needs, over the advertising
messages they receive, the amount of information they give and/or receive, and the
presentation method of such content (Hoffman & Novak, 1996).
According to Papacharissi and Rubin (2000), a key component to the uses and
gratifications theory is audience activity, which is heavily impacted by motivations. Ko,
Cho, and Roberts (2005) extended this idea to say that the four main motivations of
Internet usage are entertainment, information, social interaction, and convenience, and
that each of these motivations were strongly related with distinct types of interaction.
Several other motivations for using SNSs have been explored, including time-passing,
entertainment, relationship development, relationship maintenance, trend-following
(Hall, 2009), impression management, self-disclosure, perceived reciprocity (Park, Jin,
& Jin, 2009), communication maintenance between friends, and social compensation
20
(Barker, 2009). Upon review of these previously researched motives for using SNSs,
the majority categorize themselves nicely within the four main motivations to use the
Internet (entertainment, information, social interaction, and convenience) set forth by
Papacharissi and Rubin (2000) in a way that best suits the purpose of this study. With
SNSs mirroring many of the same functions as the Internet, and the vast majority of
researched motivations for using SNSs fitting into these four categories, it seems
appropriate to apply the motivation categories of entertainment, information, social
interaction, and convenience to SNS behavior for the purpose of this study.
Online Profiling
The manifestation of behaviors and motivations on SNSs begins with the heart
and soul of a social networking site: the construction of a profile page (Boyd & Ellison,
2007). This personal profile serves to represent the participant on a continual basis and
acts like a personal hub for each user’s participation (Goodman et al., 2008). Through
this profile creation, users provide “understandings of themselves, about themselves
and for themselves in relation to others” (Dowdall, 2009, p. 78), including in relation to
brands. This personal profile becomes a continuation of their offline social existence,
which is extended into a crafted online identity. This morphing of offline and online life
even extends to daily conversations. Conversations begun in offline situations are
continued online and vice versa. The boundaries often become blurred (Davies, 2009;
Dowdall, 2009; Goodman et al., 2008).
The idea of identity creation is not limited to basic profile creation, but also extends
to other forms of posting behavior. Willett (2009) stated that by posting content,
particularly media-related content, today’s youth can be seen as creating identities.
Although previous research has shown that identities implied by online profiles are
21
strongly related to their matching offline identities (Van Cleemput, 2009), this inclination
does not eliminate tendencies to play with it to some degree. “Social-networking sites
which combine blogs, profiles and photo and video-sharing can be viewed as cultural
resources which are used by young people as a way of performing and perhaps playing
with their identity” (Willett, 2009, pp. 55-56). In this way, users take resources available
to them and modify, reapply and recontextualize them into a “bricolage” (Levi-Strauss,
1974) that creates a new meaning and a unique identity (Willett, 2009).
Material objects, products, and brands are often used as markers of identity (Lury,
1996). This idea is part of a bigger fan culture. The Consortium (2006) labels this “Brand
Culture” and defines these as “communities of committed consumers of specific
products and services (such as BMW drivers, iPod users and Coke collectors)” (p. n.p.).
In conjunction with this, Quart (2003) proclaimed that consumer culture pegs teens as
target consumers, but that the reverse is also true in that teens are branded objects
themselves. As a result, a new landscape is being formed where products and services
are no longer simply developed and presented to the consumer in a unidirectional flow.
Instead, these fan bases are now providing feedback productions of media, text, and
interpretations (Jenkins, 2006). Products and the content that accompanies them are
now being modified uniquely by individuals who identify with them. These personal
connections and personal influencers have more impact than previously considered.
Beavis (2009) found in her research on convergence that “with respect to convergences
of ‘fan, brand and style culture’ [such as on social networking sites], brand and fan
culture had less of a role to play but ‘style culture’, and [students’] friendships and self-
concepts mattered considerably” (p. 32).
22
This trend is particularly apparent among younger users. In a PBS documentary
about youth today on SNSs, a high school boy claimed, “You need to have the Internet
on to talk to your friends [because] everybody uses it. It’s like a currency. If you [do not]
use it you’re going to be at the loss” (Goodman et al, 2008, 6:14). The online world has
become a familiar parallel universe to many people, and as such “social networking
sites are also increasingly the place where kids hash out their conflicts” (Goodman et
al., 2008, 14:46), once again blurring the lines between online and offline life. One
student interviewed about her Facebook profile replied, “Pretty much everyone has one.
It’s like a section of the Internet that is your own. Like, you can make it your personality
exactly” (13:13). One expert in the document, Pascoe, explained:
in a way, the social networking sites are a digital representation of what we think of as adolescence. So what teens are doing are going around and trying on these different identities. “I’m a goth”, or “I’m a surfer” or “I’m a punk rocker”, or “I’m a ‘this’ or ‘that’”, and the Internet’s allowed them to display that identity in a very dramatic and very distinct way.” (18:26)
This kind of play is adopted by adults as well, particularly through connectivity.
“Connectivity is increasingly part of adults’ working and social lives; like the young, they
use the Internet as a way of exploring new ways of connecting socially . . .” (Davies,
2009, p. 110).
Pascoe suggested that an opportunity for play with regard to identity is one
component of profiling (Goodman et al., 2008). Thus, profiling can be described as
more of a means-oriented practice as opposed to a goal-oriented practice that can be
applied by SNS users of all ages (Bauman, 2005). As a means-oriented practice, a user
is likely to utilize a very liquid structure, in which a profile can be modified as often and
as much as a user prefers. This means that a profile is continually evolving, just as the
23
actual user. In this case, content remains only so long as the user still feels it is
congruent with his/her own current identity to some degree (Dowdall, 2009).
The New Word-of-Mouth
Social networking sites have also impacted word-of-mouth practices. As a result of
the many-to-many pattern discussed previously (Shirky, 2009) and the practice of
identity creation using products and brands as markers of such identity, SNSs are a
prime medium for advanced word-of-mouth opportunities, which can be appealing to
marketers. Word-of-mouth is already a highly valued asset for marketers. In the SNS
environment, reach is significantly large, snowball effects thrive from multi-directional
conversations, and electronic recording of activity translates into word-of-mouth that is
more traceable and measureable than ever before (Trusov, Bucklin, & Pauwels, 2009).
Attaining this word-of-mouth advantage presents a powerful opportunity. Due to the
augmented coordination capacity of the Internet and online identity creation trends, as
mentioned earlier, users can now act as producers and consumers (Shirky, 2009). In
this way, traditional advertising efforts are being restructured to a pull strategy, which
means that companies are increasingly placing consumers in control.
Lester Wunderman, considered by many to be the creator of today’s direct
marketing, thought direct mail to be similar to a game of shooting a target in the sense
that the mail is the bullet that hits the target (selected consumers) every time
(Steinbock, 2000). The Internet presents itself to this utility as uniquely opportunistic due
to its adoption rate, ubiquity, and constant access to and ability to distribute virtually
limitless content (Arndt, 2001). The realm of consumer relationships has consequently
been altered by the moderation of the Internet. Regarding the birth of the Internet as an
advertising medium, Wunderman progressively stated, “I wanted something different. I
24
wanted a medium where the advertiser would become the target and the consumer
would become the shooter. That’s what’s happening. It is a profoundly different
marketplace” (Steinbock, 2000, p. 127).
This vision has matured with the emergence of major SNSs such as MySpace and
Facebook in the period since Wunderman’s statement in 2000. The interaction changes
accelerated by SNSs have redefined how people engage with brands by allowing the
relationships to be more personal and individualized as if the company is simply another
person to befriend (Boyd, 2006).
Wunderman stated that “with the rise of interactivity, more companies would have
to become relationship marketers and win new clients one at a time. In the process, . . .
computers would substitute human interactivity” (Steinbock, 2000, p. 129). Social
networking sites facilitate this individualized relationship marketing by attaining clients
and consumers one “friend” or “fan” at a time, creating more customized and focused
relationships with consumers who freely elect to do so. The creator and CEO of
Facebook, Mark Zuckerberg, observed this trend by saying,
[In] the last hundred years . . . the way to advertise was to get into the mass media and push your content. . . . In the next hundred years information [will not] just be pushed out to people, it will be shared among the millions of connections people have. Advertising will change. You will need to get into these connections. (Holzner, 2009, p. 1)
Discovering the factors behind SNS branded posting behavior will bring marketers one
step closer to being in the heart of these connections.
As previously stated, the consumer is therefore increasingly in control, but the
many-to-many communication pattern is also allowing brands to mold their products and
marketing efforts to better fit their consumers according to this consumer feedback
(Shirky, 2009). Previously marketers simply strove to influence and promote
25
word-of-mouth about their brands. Consumers are showing—through participation—that
they are willing to talk. They are willing to adopt brands, ideas, and concepts as their
own, as defining characteristics of themselves (Boyd & Ellison, 2007). With the
development of SNSs, marketers can still promote branded word-of-mouth, but now
have the power of the many-to-many discussions that characterize SNSs behind them
and an increased ability to compactly observe and quantify results (Trusov, Bucklin, &
Pauwels, 2009).
Boyd (2008) argued on behalf of social activists that “given the typical friend
overlap in most networks, many within those networks hear the same thing over and
over until they believe it to be true . . . [which gives] the impression that activists have
spread a message further than they have” (p. 243). This not only applies to social
activists, but also branded word-of-mouth. In addition to this increased effectiveness,
consumers voluntarily open doors on SNSs for brands to share valuable content and
promotions by electing to be “friends” or by becoming a “fan” of the brand. While
marketers’ goals are still to get people to talk about their brand, the difference is that on
an SNS those people have a place highly conducive to discussions about that brand,
and brands can even join the conversation. “Just as different national identities have
been mixed in the hybrid [of online and offline worlds], so too the realms of business
and culture are converging in novel ways” (Consalvo, 2006, p. 120). These transitions
make understanding consumer connections, motivations, and behavior even more
relevant.
Self-Congruity Theory
Two theories, the uses and gratifications theory and the self-congruity theory, help
explain what factors contribute to an individual’s branded activity on SNSs. As
26
previously discussed, the uses and gratifications theory provides a focus on how
motivating factors are likely to influence the usage of a medium to gratify specific needs
(Blumler & Katz, 1974). Thus, in alignment with this theory, it should follow that a
person’s motivation for using an SNS is likely to impact his usage behavior.
To supplement this approach in a socially influenced setting, image concepts and
perceptions can be very influential on public behavior decisions (Shaw & Costanzo,
1970). Thus, a self-congruity theory perspective could contribute insight into posting
behavior. The self-congruity theory applies self-congruity effects on self-expression to
the attitude theory by suggesting favorable predispositions toward brands that enhance
or confirm perceptions of their own self-image (Sirgy, 1986). Essentially, it is a social
cognition theory due to the fact that an awareness of “self” serves a fundamental role
and application in life experience (Rychlak, 1981). The attitude theory explains
consumers’ favorability (likes and dislikes) in connection with consumer needs, which
ultimately drives consumption behavior (Sirgy, 1986). Subsequently, self-congruity is
the match or mismatch perception associated with a consumer’s comparison between
self-image and other stimuli such as brand image, product image, or company image
(Sirgy, 1986). Moreover, Parker (2005) explained the implications of this in terms of
brand attitudes:
Self-brand congruity, the comparison of self to brand, affects brand attitudes particularly when the social signaling value of a brand is high (i.e., used in a public situation) and when symbolic, self-expressive motivations are involved. Positive brand attitudes should result as the similarity between brand image and the consumer’s self-image increases (p. 4)
Self-brand congruity as used by Parker (2005) and Sirgy (1986), is the congruity match
between self-image and brand-image. Sirgy et al. (1997) used the phrasing “typical user
of a brand” to measure brand-image instead of simple and direct references to one’s
27
image of a brand itself because “personal images of a product reflect the stereotype of
the generalized users of that product and are determined by a host of factors such as
advertising, price, and other marketing and psychological associations” (p. 229).
In his research Sirgy (1986) examined self-image and its comparison to the actual
perceptions of characteristics of the “I” or “me,” as Rogers (1959) defined “self,” as well
as the perception of the ideal self, which refers to how a person would like to view
himself. Self-image is a subset of “self” based on self-awareness that can change in
different social roles (Sirgy, 1986). This construct is a key indicator for understanding
consumer behavior (Parker, 2005). Rogers (1959) stated that individuals are motivated
by a basic and fundamental “actualizing tendency” that one hopes will, at the very least,
maintain, if not enhance a person’s self-image. Furthermore, Franken (1994) indicates
the structural basis on how the concept of self becomes actualized:
There is great deal of research which shows that the self-concept is, perhaps, the basis for all motivated behavior. It is the self-concept that gives rise to possible selves, and it is possible selves that create the motivation for behavior. (p. 443)
Thus, both the uses and gratifications theory and self-congruity theory provide
rounded reasoning for a resulting conclusion that the more a person thinks that a brand
is congruent with his own self-images, the more he will be motivated to behave in ways
that reinforce those self-images on SNSs for various gratification purposes. If a person’s
natural tendency is to self-actualize according to his motivations as Rogers (1959)
claimed, it seems probable that such a person would display these behaviors publicly
via his own profiles and through social interaction on SNSs.
28
Hypotheses
This study expanded upon previous theoretical models involving antecedents of
media usage, particularly SNSs, and congruity between an individual and a brand within
such a medium. The constructs here are dependent upon familiarity and loyalty with the
brand and the medium. Familiarity and loyalty are then used to shed insight on a
person’s motivation and likelihood to integrate a brand on his SNS profile as a
personally identifying characteristic. The operational definitions in this case are
important to clarify due to the colloquially interchangeable use of terms such as “online
communities,” “social networking utilities,” “online social networks,” and “social
networking sites.” For the purposes of this study “social networking sites” (SNSs) will be
the term used. Accordingly, the concepts to be measured in this theoretical
development include self-brand congruity, brand-SNS compatibility, brand attitude, SNS
usage motivations, and finally the resulting brand adoption behavior to one’s SNS
activity.
A brand, according to the American Marketing Association is a “name, term, sign,
symbol, or design, or a combination of them, intended to identify the goods and services
of one seller or group of sellers and to differentiate them from those of competition”
(Keller, 2008, p. 2). However, Keller (2008) widened this definition to a more universal
industry concept by stating that a brand is also “something that has actually created a
certain amount of awareness, reputation, prominence, and so on in the marketplace”
(Keller, 2008, p. 2). Though he limited this brand development to the marketplace,
people in general are increasingly branding themselves through the phenomenon of
online profiling; creating the awareness, reputation, and prominence that Keller (2008)
referred to. It is this consumption need of self-expression that often drives consumers to
29
“purchase brands (e.g., Gucci and Rolex) that communicate a particular image or social
role” (Parker, 2005, p.10). Thus, they are creating a brand for themselves often through
a selected compilation of other brands in the marketplace. Some insight into this self-
expression practice can be provided by the self-brand congruity concept.
Self-brand congruity is based upon attitudes regarding oneself and a brand, or
rather, the typical user of a brand. Attitudes are typically researched in marketing and
advertising because they have shown to perceptively predict consumer behavior
(Mitchell & Olson, 1981). The core influences of behavior are the cognitive building
blocks, or beliefs, which are considered to be fundamental in developing attitudes
(Fishbein & Ajzen, 1975).
Thus, an individual’s beliefs are likely to be highly translated into his introspective
attitudes and their attitudes toward brands. Since attitudes have been shown to mold
behavior in past consumer behavior research, it is not unreasonable to presume that
these beliefs and attitudes can be translated into measureable, personally identifying
actions (Parker, 2005). Self-brand congruity is defined by this perceived alignment of
attitudes toward a brand and the beliefs that make up their identity. For the purposes of
this study, the resulting behavior likelihood is defined as the likely or unlikely adoption of
a brand to be presented in any fashion on an individual’s Facebook activity or profile
page, referred to as “brand adoption behavior likelihood” in this study. This behavior is
proposed to be indicative of a person’s actual and ideal self-brand congruity with the
particular brand.
• Hypothesis 1: Self-brand congruity (both actual and ideal) will be positively related to the likelihood of SNS brand adoption behavior.
30
It is possible, however, that self-brand congruity may be present, yet not be
exhibited on an SNS because the brand or product category may not be considered
suitable for a public persona so openly broadcasted. For example, a product may not be
deemed as relevant, applicable, appropriate, or natural for Facebook topics. Thus, the
individual may not consider a brand or product category compatible with Facebook.
This possible lack of compatibility must be addressed in an effort to make this research
more reliable. Thus, the more an individual considers a brand as being relatable and
suitable for Facebook, the more likely that person will be to display such accordance.
• Hypothesis 2: Brand-SNS compatibility will be positively related to the likelihood of SNS brand adoption behavior.
As previously stated, in order for a brand to be selected by the individual for such
self-expressive profiling, it is important that he have a favorable consumer attitude
toward it. The same could also be true for the reverse situation. If a person has a strong
unfavorable attitude toward a brand, he may consider that anti-congruity to be an
integral part of his personal characteristic set as well. With this assumption in mind, the
current study will assumed this association to be true and executed evaluations solely in
positive terms.
Parker (2005) declared, “In general, consumers tend to favor brands and products
that satisfy their needs and wants better than competitive choices” (p. 10). This
statement implies a working application of the uses and gratifications theory. The four
primary motivations for Internet usage laid out by Kaye and Johnson (2001) and
Papacharissi and Rubin (2000) were information, convenience, entertainment, and
social interaction. As previously discussed, these four fundamental Internet usage
31
motivations can be applied more specifically to the use of SNSs to derive insight on
their effect on behavior in this new context.
Within these four motivations, previous studies have indicated that media with
higher interactive value or usage value (such as entertainment or information), tend to
induce higher motivation in consumers to use them (Ko, Cho, & Roberts, 2005).
Concurrent with this finding, the entertainment and information motivations should
inspire a positive influence on brand adoption behavior likelihood.
• Hypothesis 3: An entertainment SNS motivation will positively interact with both actual and ideal self-brand congruity in predicting SNS adoption behavior likelihood.
• Hypothesis 4: An information SNS motivation will positively interact with both actual and ideal self-brand congruity in predicting SNS adoption behavior likelihood.
Furthermore, social interaction is highly impacted by perceived social
consequences. Associations are therefore carefully expressed. One study found that
when users express their political affiliations, they are typically prompted to justify their
choices (Wu, 2009). Such accountability and social judgment would likely inflict a sense
of reserve and prudence in freely admitting to associations unless the individual feels
very strongly about them. Moreover, if a person is using an SNS out of convenience
motivations, it seems unlikely that the person would exhibit the effort of associating
himself with a brand in extended behavior or exhibiting critical evaluation involved in
justifying such associations (Wu, 2009). Therefore, both of these motivations can be
expected to stimulate a negative influence on the addition of a brand to a person’s SNS
posting behavior. Thus the following research questions arise:
• Research Question 1: Do social interaction SNS motivations negatively interact with actual and ideal self-brand congruity in predicting SNS brand adoption behavior likelihood?
32
• Research Question 2: Do convenience SNS motivations negatively interact with actual and ideal self-brand congruity in predicting SNS brand adoption behavior likelihood?
33
Figure 2-1 Overview of the study.
Self-Brand Congruity
Brand-SNS Compatibility
Adoption Behavior Likelihood
SNS Usage Motivations
Entertainment (H3)
Information (H4)
Social Interaction (RQ1)
Convenience (RQ2)
Interaction Effect (H3, H4, RQ1, RQ2)
(H1)
(H2)
34
CHAPTER 3 METHOD
The purpose of this study was to explain branded posting behavior on SNSs. Such
answers would ideally provide brand management answers for how to excel using
social media as a pivotal resource. The goal was to gain insight from active SNS
participants and consumers that could unlock new approaches to brand management.
This study executed this goal primarily by looking at self-brand congruity measures, but
also by observing participants’ motivation(s) for posting on an SNS as an interacting
effect. The data for this research were collected using an online survey with self-
administered questionnaires. The variables measured in reference to the leading SNS,
Facebook, included 1) familiarity with Facebook and the brand, 2) attitudes toward
Facebook and the brand, 3) current Facebook behavior, 4) image perceptions regarding
self-brand congruity as well as brand-SNS compatibility, and 5) their resulting adoption
behavior likelihood.
Research Design
This study employed an online survey to fulfill its research purpose. The
questionnaires for this survey were available for completion on www.surveymonkey.com
from May 15, 2009 through May 26, 2009 (Appendix B). Sample participants received
an invitation to complete a questionnaire, which directed them to the appropriate link
(Appendix A).
In completing the questionnaire, each participant was asked to evaluate himself,
an individual brand, and Facebook on qualities such as familiarity, image perception,
self-congruity, and compatibility between the brands and Facebook. These constructs
were assessed using Likert scales with five levels. Variables were typically coded with
35
the most positive response equaling one, and more negative responses were assigned
incrementally increasing values. The data for each questionnaire were coded and
recorded into a SPSS (formerly PASW) database. The data were then analyzed for
correlations with actions taken in reference to the brand on Facebook through
regression analysis. Descriptive and frequency analyses were also executed to
supplement these observations.
Participants
The sampling method employed in this case was a non-probability purposive
convenience sample. Purposive sampling depends on the researcher to select
participants—within their convenience and judgment—who most appropriately represent
the targeted sample (Babbie, 2007). To obtain a valid sample of SNS users within the
scope and constraints of this study, purposive sampling was combined with the
snowball sampling method, where “a researcher collects data on the few members of
the target population he or she can locate, then asks those individuals to provide the
information needed to locate other members of that population whom they happen to
know” (Babbie, 2007, p. 185). This combination was especially appropriate for this study
because it asked the population, Facebook users, to participate in a way that mirrored
the kind of SNS behavior in which the study was interested. Thus, to fulfill the purpose
set out for this research, a total of 151 respondents participated in the study. Of these
151 participants, 74 responses qualified for inclusion in analysis of this survey. These
participants were registered users of Facebook and coffee drinkers who were at least
18 years of age.
36
Data Collection and Procedure
Surveys are used for studies with descriptive, explanatory, and exploratory
purposes where the units of study are typically individuals as opposed to groups or
interactions (Babbie, 2007). Babbie (2007) also stated that “survey research is probably
the best method available to the social researcher who is interested in collecting original
data for describing a population too large to observe directly,” and it can be an
“excellent vehicle for measuring attitudes and orientations in a large population”
(p. 244). Privacy limitations of SNSs and the vast population of individual SNS users
eliminate the possibility of observing the entire population. As a result, survey research
that relies on a representative sample to characterize the population was the most
effective method to provide the attitudinal and behavioral insight desired in this study.
This method was utilized to gain exploratory insight on participants due to the
succinctness of both the method and the medium. Focus groups are typically most
appropriate for exploratory inquiries where limited research on the specific subject in
question has been done previously and inductive reasoning is required. However, they
carry a risk of the chameleon affect occurring. The chameleon affect is the
“nonconscious mimicry of the postures, mannerisms, facial expressions, and other
behaviors of one's interaction partners, such that one's behavior passively and
unintentionally changes to match that of others in one's current social environment”
(Chartrand & Bargh, 1999, p. 893).
Instead, an online survey allowed for much of the same questions to be asked as
in a focus group or personal interview, but with distinct and definitive answers that
mirrored the SNS medium in question. Thus, minimal, if any, valuable insight was lost.
The objective of this exploratory study was to obtain measurable and meaningful
37
information from participants by identifying motivational, attitudinal, and behavioral
patterns.
To provide sound explanation for the sampling method chosen, it is important to
clarify why Facebook was the social networking site chosen for this research. As
explained earlier, SNSs often target specific groups of users. Therefore, in selecting a
manageable representation of SNSs, the scope of this research was narrowed down to
Facebook for a variety of reasons. First, Facebook is an SNS with one of the highest
number of members and the most registered activity by facilitating interaction and
sharing of news and media between individuals and groups (Lopez, 2008). Beyond this
top-ranked popularity, Facebook’s users are fairly diverse demographically, but are
highly saturated among college students due to its initial exclusivity as a utility solely for
college students (About Facebook, 2004). Facebook’s qualities position it as most
aligned with the available convenience sample population, but also most likely to
provide respondents who are familiar with the specific social utility. The sample chosen
mirrors the users of Facebook.
The first execution of this sampling was done via an invitation requesting
Facebook users’ participation in the study which was posted directly on Facebook.
Facebook viewers of the invitation were encouraged to repost it and pass it along to
others within their network. This invitation explained what the study regarded, as well as
what participation meant for them in terms of time, effort, and confidentiality.
A second purposive convenience sample was used in conjunction with the first
sample. The sample population for this group was college undergraduate students
between the ages of 18 and 25, enrolled in Journalism and Communication classes at
38
the University of Florida for the 2008-2009 academic year. The same invitation to
participate was distributed to students in three classes within the University of Florida’s
Mass Communication department. Students participated voluntarily and, at the
discretion of each class’s professor, the classes were given the opportunity to earn
course credit for the completion of the questionnaire. To keep the questionnaire within a
reasonable length, one questionnaire was dedicated to each of the two brands used.
Participants were directed, according to last name, to one of the two questionnaires to
complete.
Similar to the carefully chosen SNS for this research, the brands observed in the
questionnaires were also well considered. Without establishing a level of familiarity with
a brand, answers to subsequent questions regarding that brand would carry little
meaning, if answers were possible at all. Thus, familiarity was an important founding
construct of this study, and finding brands with which participants were highly familiar
improved the research validity. As a result, two brands were chosen: Starbucks, a
well-known brand internationally, and Dunkin’ Donuts, a well-known coffee provider in
the United States, but with a slightly different consumer base. The product category of
coffee distributors was chosen for its prominence among college students, as it remains
a staple in many college students’ lives. It was also ubiquitous enough that regardless
of coffee consumption levels, the sample recruited would likely have attitudes and
beliefs regarding the brands within the category. In fact, in a study about the differences
between in-home coffee drinking and in-store consumers, 46% of respondents ordered
non-coffee products, which implies additional reasons for developing attitudes toward
39
coffee shops than the mere liking of coffee (Henson, 2007). Furthermore, according to
Parker’s (2005) pretests, Starbucks was among the top 25 publicly consumed brands.
A pre-test was executed to determine the most appropriate brands for the study.
Brands tested in the pre-test included two national companies, Starbucks and Dunkin’
Donuts, and three local companies in Gainesville, Florida, Barnie’s Coffee & Tea Co.,
Maude’s Classic Café, and Lollicup Coffee & Tea. The pre-test was very brief and
included questions evaluating familiarity, attitudes, self-brand congruity, and brand
compatibility with Facebook. Starbucks and Dunkin’ Donuts proved to be the most
appropriate brands for this study. They both had similar high levels of familiarity, yet
attitude, congruity, and compatibility answers showed that they did not have the same
group of consumers.
Further analysis for the actual study included a review of each brand’s set of
responses and entering the data into PASW. Frequency analysis and simple summary
statistical tests were applied to each brand group individually, as well as the combined
sample data, to provide a descriptive sample profile and a summary of existing
Facebook activity. Subsequently, several comparisons of means analyses were
executed to establish significant differences between the two brands regarding
demographics, behaviors, and psychographics. Among the variables assessed in these
comparisons were independent variables of brand attitude, actual and ideal self-brand
congruity, and brand-Facebook compatibility. Next, bivariate correlation and multiple
regression tests were performed to determine whether to accept or reject each
hypothesis, as well as to explore possible answers to the research questions. From this
synthesized information, the statistical tests were evaluated on whether the data
40
provided sufficient answers regarding the correlation between each independent or
interaction variable (explained in more detail in the following sections) and the
dependent variable—adoption behavior likelihood on Facebook. The results of this
research are detailed in Chapter 4.
Measures and Instrument
The materials required for this study included a questionnaire (see Appendix B), a
computer with Internet access, and an invitation to complete the questionnaire (see
Appendix A). The invitation was brief, describing the general context of the study and
participation, as well as providing the link where the students could find and complete
the questionnaire.
As previously stated, this study included two sets of questionnaires: one set of
questions regarding Starbucks and the same set of questions regarding Dunkin’ Donuts.
This split was done to shorten the length of the questionnaire with the intention of
increasing completion rates. Furthermore, within the questionnaires, it was important to
set background context with a list of questions regarding the participants’ behavior
tendencies in their current Facebook usage. The questionnaire also targeted topics
about attitudes, congruity, compatibility, and adoption behavior. Some questions asked
participants to role play in a given situation; other questions simply asked participants to
what degree they agree or disagree with an item.
Each questionnaire consisted of eight sections. These sections aimed to establish
1) participant qualification; 2) motivations for using Facebook; 3) familiarity with, usage
level of, and attitudes toward Facebook; 4) existing user habits on Facebook; 5)
familiarity and attitudes toward the brand; 6) self-brand congruity and brand-Facebook
compatibility; 7) adoption behavior likelihood of the brand; and 8) a demographic profile.
41
More specifically, the survey questionnaire began with a series of qualification and
background questions regarding Facebook. To qualify to complete the entire
questionnaire, participants were required to be coffee drinkers, and also have a
registered Facebook account. Question 3 asked for the date of initial membership with
Facebook to discover the possible level of expertise and give more meaning to activity
on these sites on the basis of membership length.
Questions 17 through 28 focused on the participants’ level of familiarity with,
usage of, and attitudes toward Facebook. This section consisted of 5-point semantic
differential questions based on questionnaires by Douglas (1999) and Kang (1999). Of
these differential questions, Questions 19 through 22 ascertained how active
participants are on the site. Question 21 indicated participants’ approximate total
activity time on Facebook per week. Although this question was separated into two
variables for hours and minutes in data collection, the two were converted into a single
hourly measurement and reported in hours. Questions 29 through 38 established the
user’s current habits on Facebook and operationalize the current Facebook behavior
variable. In doing so, the results of the 10 questions for each participant were summed
to create a current Facebook behavior index where a maximum score of “44” indicates
the least behavior and familiarity and a minimum score of “10” indicates the most
intense behavior practices. This measure helped to give proper significance to results of
the research. For example, if a participant starts out at a high level of activity, it shows
that this individual is willing to display information about himself. The same idea applies
to the other extreme.
42
Independent Variable
Two official independent variables were established in this study. These were 1)
self-brand congruity (Hypothesis 1) where both actual and ideal self-brand congruity
were observed, and 2)brand-Facebook compatibility (Hypothesis 2). However, an added
interest was observed in brand attitude as a corroborating independent variable. This
observation was implemented to confirm the theoretical progression from brand attitude
to self-brand congruity and finally, adoption behavior as laid out by previous research.
Brand attitude and differences between brand groups were explored for a
preliminary assessment of the appropriateness of measures with alignment to previous
research and an assessment of different brands’ influences on self-brand congruity and
brand-Facebook compatibility. Confirming a lack of significant differences between
brand groups lends more justification to examining the variables of self-brand congruity
and brand-Facebook compatibility outlined for Hypothesis 1 and Hypothesis 2. On the
other hand, confirmed significant differences between brand groups would indicate a
need to keep brand influence in mind when considering the results of these tests.
Furthermore, to rule out significant differences between brands would mean that both
brand samples could be evaluated together as one combined sample. This would add to
the overall sample size and reliability of the research.
To give further depth to the evaluation of the independent brand groups, familiarity
with the brand was assessed briefly, and brand attitude was considered to a further
extent. To establish familiarity with the brand and measure brand attitude, the chosen
brands, Starbucks and Dunkin’ Donuts, were introduced in each parallel questionnaire
by asking semantic differential questions (Questions 39 through 43). Familiarity was
addressed with a single semantic differential on a 5-point scale from “1” (“very familiar”)
43
to “5” (“very unfamiliar”). Furthermore, attitude toward the brand was also measured via
four semantic differential questions (40 through 43) rating the following four qualities for
the brand: 1) good/bad, 2) pleasant/unpleasant, 3) favorable/unfavorable, and 4)
likable/unlikable. These questions were also adopted from questionnaires by
Papacharissi and Rubin (2000), Douglas (1999), Kang (1999), and Parker (2005). This
method of evaluation is commonly used to measure brand attitude (MacKenzie, Lutz, &
Park, 1989). The results of the four questions were averaged to create an index for
operationalization of the brand attitude variable with a reliable Cronbach’s alpha
calculated in previous research as .83.
Next, the independent variable of self-brand congruity was addressed in the latter
half of the questionnaire. This section linked the individual, the environment, and the
brands by asking questions regarding the participants’ image perceptions and level of
congruity with the brands within the context of Facebook. The initial questions for this
section (questions 44 through 48) were adopted from the global self-congruity scales
method proposed by Sirgy (1997) and further executed in Parker’s self-congruity
research (2005). This method first determined brand user imagery descriptors and
subsequently measured self-brand congruity based on the descriptors cataloged. The
respondent was first asked to provide two adjectives to describe the typical user of the
brand in question. Next, the respondent was asked questions to establish the level of
congruity he felt between his own actual and ideal self-images and the descriptors he
just used to describe the typical user of the brand as follows:
Take a moment to think about [Brand x]. Think about the kind of person who typically uses the brand. Imagine this person in your mind and then describe this person using one or two personal adjectives to describe the typical user of the brand. Now indicate your agreement or disagreement to
44
the following statements: The typical user of [Brand x] is consistent with how I see myself (actual self congruity); The typical user of [Brand x] is consistent with how I like to see myself (ideal self congruity); The [Brand x] brand is compatible with Facebook (brand-Facebook compatibility).
As labeled above, the first of these questions is intended to measure actual
self-congruity, while the second question is intended to measure ideal self-congruity.
The congruity indicator scores collected from both of these questions were averaged to
create a solitary self-brand congruity index to simplify analysis and discussion, provided
no significant difference occurred between the actual and ideal congruity responses. If
significant differences occurred, however, actual and ideal congruity were treated
separately as individual indicators of self-brand congruity. This method measures
self-brand congruity directly, as opposed to performing separate evaluations of self-
image and brand (or product-user) image (Sirgy et al., 1997). The method proposed by
Sirgy et al. (1997) also took a holistic and global perspective on the measurement by
allowing participants to conjure up their own perception descriptors for typical brand
users instead of presenting them with predetermined image perceptions to evaluate
congruity (Sirgy et al., 1997). High or low self-brand congruity was determined by the
level to which the participant indicated he perceived a match (or mismatch) between his
own self-image (actual and ideal) and the perceived image of the typical user of the
brand. In this study, high self-brand congruity was indicated with scores closer to 1,
while scores closer to 5 indicated low self-brand congruity.
Brand-Facebook compatibility, another independent variable, was subsequently
measured as part of this set by asking the participant to indicate his level of agreement
or disagreement that the brand is compatible with Facebook. This measurement used a
solitary 5-point Likert-scale ranging from “1” (“strongly agree”) to “5” (“strongly
45
disagree”). The identical questions were asked for both Starbucks and Dunkin’ Donuts
on their respective questionnaires for each of these independent variables.
Dependent Variable
The dependent variable in this study was brand adoption behavior likelihood on a
Facebook user’s posting activity. This variable was examined in the final portion of the
questionnaire with two questions. Questions 49 and 50 asked the respondent to role
play in various Facebook situations. First, in Question 49, the respondent was simply
asked whether he would add the brand to his profile to which he must answer his level
of agreement on a 5-point scale which was coded from “1 (“no, definitely not”) to “5”
(“yes, definitely”). Thus, a higher number indicated a higher likelihood of adoption
behavior.
Since research is limited on specific posting behavior intention, this first dependent
variable question was based on measures of intended voting behavior. For example,
the following question was utilized to determine candidate preference in a study that
examined the impact of negative political television commercials, “If you were voting in
this election, after seeing this commercial, would you vote for the sponsoring
candidate?” Respondents were then asked to indicate their selection on a scale ranging
from “no, definitely not” (-3) to “yes, definitely” (+3) (Tinkham & Weaver-Lariscy, 1993).
As reinforcement to this measurement, the next question asked about more
detailed posting behavior decisions. A list of 18 possible brand adoption behaviors, with
positive, neutral, and negative brand attitudes, were presented to the respondent, along
with an “Other” option. Ten of these items were denoted as positive actions regarding
the brand, five were considered neutral, and three were considered to be negative
actions. Neutral and negative actions were included to provide a more well-rounded
46
pool of possible actions reflective of typical Facebook behavior. This list of 18 behaviors
was chosen because these behaviors provided a possible range typical of Facebook
users, which gave the opportunity for a user to publicly express or display a brand in his
Facebook activity. The participant was asked to check all the actions he was likely to
take. Since this study aimed to identify positive effects on behavior, the 10 positive
action items were the focus for data analysis. To operationalize this dependent variable,
a value of “1” was assigned if the participant indicated he would likely take that action,
and a value of “0” was assigned if the action was not selected. Responses to the 10
positive items were tallied to create an aggregate positive action score. This aggregate
score from “0” to “10” was added to the score from the previous question to create a
total adoption behavior likelihood index. This index had possible scores from “1” to “15”
as a result. In this case, the higher the adoption behavior likelihood index score, the
more likely a person was to take action regarding the brand in his Facebook activity.
Question 50 was originally intended to modify the model developed by Curry
(2004) which measured decision-making bases in which each answer was assigned a
positive, negative, or zero value. These values were aggregated to create four
composite scores for each of the items measured in Curry’s research (Curry, 2004).
This aggregated scoring was adapted to the current study with a few changes. The
main modification made to this model was that essentially one item was examined in
this section of the questionnaire and multiple response options were not mutually
exclusive. Thus, answer options were increased and one composite score was created
as a result of this modification. This model was chosen because users could perform
more than one of the proposed behaviors simultaneously, any of which could potentially
47
express varying attitudes toward the indicated brand on a positive-negative continuum.
As with Curry’s model, positive and negative aggregate scoring could be applied in
future research to determine a correlation between negative brand attitude, self-brand
congruity, and behaviors with negative attitudes, or to observe inaction tendencies as
opposed to posting with positive or negative attitudes toward a brand.
Motivation Interaction Variables
The interaction variables in this research explored the interacting effects of various
motivational reasons for using Facebook. A set of 13 questions addressed the four main
motivations set forth by Papacharissi and Rubin (2000). These motivations were
information (Questions 4 through 6), convenience (Questions 7 through 9),
entertainment (Questions 10 through 13), and social interaction (Questions 14 through
16). Each of the four motivations had three questions dedicated to its measurement
except for entertainment which included four questions in the set. The participant was
asked to indicate his level of agreement with these 13 statements on a 5-point scale
from “1” (“strongly agree”) to “5” (“strongly disagree”). Results for each question within a
motivation set were averaged to create an index score for that motivation. As a result,
lower motivation indexes indicated that a participant was more motivated by that usage
purpose. This model indicated sufficient reliability with a collective Cronbach’s alpha
score of .78 (Ko, Cho, & Roberts, 2005). Motivational indices were then multiplied by
the self-brand congruity indices to create an interaction variable for each motivation.
Self-brand congruity indices, individual motivation indices, and interaction variables
were all entered into multiple regression analyses for each of the four motivations
examined in Hypothesis 3, Hypothesis 4, Research Question 1, and Research
Question 2.
48
CHAPTER 4 RESULTS
Data Analysis
Sample Profile
This study recruited 151 participants for the online survey. However, due to the
nature of the study and the qualifying measures required to produce valid data, 77
respondents were eliminated from the final analysis of results. Respondents were
removed for two reasons: 1) respondents’ voluntary incompletion of the survey and 2)
respondents’ not drinking coffee and/or not having a registered Facebook account.
Therefore, many respondents were removed because they did not fit one or both of
these qualifications, and thus were unable to answer the questions relevant to the core
purpose of this study. Taking these eliminations into account, the final valid sample
includes 74 respondents. Of these 74 participants, 41 took the questionnaire with
questions regarding Starbucks, and 33 took a questionnaire with questions regarding
Dunkin’ Donuts.
Table 4-1 displays the summary description of the Starbucks, Dunkin’ Donuts, and
total sample. The Starbucks and Dunkin’ Donuts groups were not significantly different
(Table 4-1). Mostly female, participants ranged in age from 18 to 41 with the majority 18
to 24. These statistics are in alignment with Facebook’s original target market age range
of 18 to 24. Just over 80% of the participants were in the process of earning an
undergraduate degree or had already completed a degree. More than half of the
respondents were Caucasian.
49
Facebook Tendencies/Activity
Means for familiarity, acceptability and likelihood to visit Facebook were high for
both groups (Table 4-1), with no significant difference between groups. Likewise, with
no difference between groups, 85.2% of the respondents use Facebook at least on a
daily basis (once per day plus multiple times per day). Participants’ activity time on
Facebook averaged 6.09 hours per week.
Participants’ level of public expression on Facebook was measured for direct
(Table 4-2) and indirect expression of information (Table 4-3). The direct expression
informational categories typically completed on Facebook included Basic Information,
Personal Information, Contact Information, and Educational and Work Information.
Among these direct expression behaviors, the most completed and detailed section was
the Basic Information section with 25% of the respondents stating it to be fully
completed, whereas respondents typically indicated they provide very limited
information in the Personal, Contact, and Work and Education Sections. This would
imply a general sense of privacy and inaction with respect to directly providing
descriptive details about themselves. This also supports the tendency for members to
primarily use SNSs to maintain current relationships (with those who would likely
already have such information such as contact, work, education and personal
information) rather than to search out new relationships (Boyd & Ellison, 2007).
The indirect expression of information was represented by a display of
associations with other items that they either searched for or were prompted to add by
other Facebook users. These items included Groups, Fan items, Bumper Stickers, Gifts,
and Other items. Groups came out as the most common behavior by far for each of the
three samples with 91.9% of the combined sample noting that they had elected to
50
become a member of at least a few groups, if not many groups. Gifts sent from other
Facebook users, Other Items, and Fan Items also presented a majority of users who
reported they showed a few of these items on their profile. Bumper stickers found by
the participants themselves exhibited the least common behavior with a majority
choosing to display none in that category.
Overall Facebook behavior was calculated by summing each respondent’s current
profile behavior with a maximum possible score of “44” (indicative of lesser activity) and
a low possible score of “10” (indicative of greater activity) (Table 4-3). This
measurement resulted in total current activity scores between 23.0 and 24.0 for both
Starbucks and Dunkin’ Donuts individually, as well as for the combined sample. On the
scale in this study from most active (“10”) to least active (“44”) these scores fell on the
more active side of the median, demonstrating moderate to high behavioral tendencies
for each group. No significant differences occurred between groups for total current
Facebook behavior or any of the individual behaviors.
Brand Attitudes, Congruity, and Compatibility
Consistent with previous use of the brand attitude measures and suggested
acceptable levels, between .7 and 1.0 (Davis, 1997), the items used for the brand
attitude index produced a Cronbach’s alpha of .95 overall, .92 for the Starbucks group,
and .97 for the Dunkin’ Donuts group (Table 4-4). While the data showed no significant
difference of familiarity between the brands, the attitude toward Starbucks was
significantly more positive than the attitude toward Dunkin’ Donuts. Self-brand congruity
components and brand-Facebook compatibility also returned significant differences
between Starbucks and Dunkin’ Donuts (Table 4-4).
51
Although no significant differences occurred between actual and ideal self-brand
congruity ratings for Dunkin’ Donuts, Starbucks’s actual self-brand congruity ratings
were significantly greater than Starbucks’s ideal self-brand congruity ratings. Since at
least one of the two observed brands revealed significant differences between these
self-brand congruity components, both actual and ideal self-brand congruity measures
will be used as separate variables in each sample group for the remainder of this
study’s statistical analysis.
In addition to significant differences between actual and ideal self-brand congruity,
significant differences were also present between groups. Starbucks’s respondents
rated Starbucks as significantly more congruent with their actual and ideal selves than
Dunkin’ Donuts’ respondents rated their actual and ideal selves. Lastly, Starbucks was
perceived to be significantly more compatible with Facebook than Dunkin’ Donuts.
Since participants also had significantly more positive attitudes toward Starbucks than
Dunkin’ Donuts, this could be indicative that if a person feels negative associations
toward a brand, he feels more strongly that the brand is not appropriate for SNS
discussion or favorable in making positive social impressions and vice versa.
Independent, Dependent, and Interaction Variables
Table 4-5 presents a summary of this study’s key variables, means, and reliability
measures. For multiple item indicators, Cronbach alphas were reported, and the
Pearson correlation coefficient was reported for the dual components (actual and ideal)
of self-brand congruity. Compatibility was a single item measure. While most of the
alphas met the acceptable minimum (.70) (Davis, 1997), several did not (Convenience
and Social Interaction). However, given the sample sizes’ link to unstable correlations, it
was decided that it was better to include the scales rather than delete them—keeping
52
the low alphas in mind when the data were analyzed. Significant differences between
brands were observed for each of the independent variables and for one motivational
variable (Information). The other three motivations showed no significant difference
between brands.
Dependent Variable: Adoptive Behavior
The dependent variable, adoption behavior likelihood, consisted of two measures.
The first was a single item scale asking the respondent to indicate if he would either
visually or verbally add the brand to his Facebook (“1” = “no, definitely not” to “5” = ”yes,
definitely”) (see “Likely of Adding” in Table 4-7). The second measure, which was added
to the first measure to create the calculated dependent variable, consisted of the sum of
all the positive actions a participant indicated he was likely to take regarding the brand
on Facebook. These two behavioral measures were significantly and moderately-
strongly correlated for the combined sample and for each brand individually (Table 4-6),
indicating these were appropriate measures.
Table 4-7 shows descriptive statistics and frequencies for each of the two
measures used to calculate the dependent variable, adoption behavior likelihood. Total
positive actions were significantly higher for Starbucks than for Dunkin’ Donuts. In a
more detailed look at each of the positive actions, two action items displayed
significantly higher results for Starbucks than for Dunkin’ Donuts. These action items
included sending a brand logo gift to someone and adding a bumper sticker of the brand
logo that was sent by another user.
The three actions most frequently indicated for the combined group were to
“accept and display a brand logo gift that someone sent you,” “write/post/display
something positive about a brand in your profile or status,” and “become a fan of the
53
brand on Facebook.” These items also appear within the most frequent responses for
both the Starbucks and Dunkin’ Donuts sample groups. The action least frequently
indicated was to actively search out and join a group with a positive attitude toward the
brand. This item was followed closely by the act of searching for the brand’s Facebook
profile and more information regarding the brand. The results in Table 4-7 show that
participants were least likely to fulfill action items requiring a participant to actively go
out of his way to seek out the brand or media feature.
Hypothesis Testing
This study explored possible relationships among a number of variables within the
concepts of consumer behavior, social psychology, brand management, and new media
communications. The constructs at the heart of this study, which have been briefly
explored thus far, included self-brand congruity, brand-SNS compatibility, and the four
SNS motivation variables of entertainment, information, social interaction, and
convenience. For the scope of this study, the best way to uncover answers to the
hypotheses between these variables was by executing regression analyses to
determine relationships.
• Hypothesis 1: Self-brand congruity (both actual and ideal) will be positively related to the likelihood of SNS brand adoptive behavior.
First, a bivariate correlation test was performed examining indexes for actual and
ideal self-brand congruity, current Facebook behavior, brand attitude, and adoption
behavior likelihood (Table 4-8). Adoption Behavior Likelihood is inversely coded from
the remaining variables. Negative correlation values are therefore indicative of positive
relationships.
54
This bivariate correlation examination highlighted a few key observations. First,
current Facebook behavior showed no significant correlation, and thus no relationship
with either self-brand congruity or brand attitude. Furthermore, though a significant
correlation was observed between current behavior and adoption behavior likelihood for
the combined sample, evidence showed only a weak positive relationship (R = -.27).
These apparent discrepancies suggest that positive attitudes and highly perceived
self-brand congruity may be highly correlated with likely adoption behaviors indicated by
participants, but this, in turn, may not necessarily translate strongly into fulfilled
behavior.
Second, brand attitude displayed a significant moderate relationship with actual
self-brand congruity, ideal self-brand congruity, and adoption behavior likelihood for all
three sample groups. This result further confirms previous research and exemplifies
these variables as good measures. Third, since brand attitude and adoption behavior
likelihood had a significantly moderate relationship, it reasonably follows that actual and
ideal self-brand congruity would also illustrate a significant moderate relationship with
the dependent variable. This was, in fact, the case for the combined sample and
Starbucks. Dunkin’ Donuts, however, showed only ideal self-brand congruity to have a
significant moderate relationship with adoption behavior likelihood. The important
overall take-away from these observations was that as brand attitude and both actual
and ideal self-brand congruity increased positively, adoption behavior likelihood also
increased, revealing a significant and positive moderate relationship.
With these confirmations in mind, Hypothesis 1 further examined the components
of self-brand congruity using bivariate correlations (Table 4-9). For the purpose of this
55
test, actual and ideal self-brand congruity measures were recoded such that “1” =
“strongly disagree” and “5” = “strongly agree”, in order to have positive correlations
depict a positive relationship and vice versa. Also, since preliminary statistical tests
showed significant differences between groups for both actual and ideal self-brand
congruity, bivariate correlation tests were run for both the Starbucks and Dunkin’ Donuts
samples in addition to the combined sample.
Three separate correlations were examined with regard to both actual and ideal
self-brand congruity. The first set looked at participants’ intent to add the brand to their
Facebook profile either verbally or visually. This measurement showed significant
moderately positive correlations with ideal self-brand congruity for Starbucks, Dunkin’
Donuts, and the combined sample. But for actual self-brand congruity this was only the
case for the Starbucks and combined sample.
The second set observed the correlation with the sum of positive actions which
participants indicated they would likely take. Both actual and ideal self-brand congruity
revealed significant, slightly weaker moderate correlations than in the previous
observation set for both Starbucks and the combined sample. However, Dunkin’ Donuts
was not significant at the p < .05 level for either actual or ideal self-brand congruity this
time.
The third set calculated dependent variable for this study—adoption behavior
likelihood—was compared with the self-brand congruity components. As previously
stated, adoption behavior likelihood was calculated by adding the score of the user’s
intent to add the brand to his Facebook activity (min = 1, max = 5) (Question 49 of the
questionnaire) and the sum of positive actions (min = 0, max = 10) which the participant
56
indicated he would take regarding the brand on Facebook (Question 50 of the
questionnaire). The bivariate correlation results were similar to the first set regarding
intent to add the brand to Facebook activity. Both actual and ideal self-brand congruity
correlations, with adoption behavior likelihood were significantly positive with moderate
strength for both Starbucks and the combined sample. Dunkin’ Donuts however,
returned ideal self-brand congruity as significantly and moderately correlated with ideal
self-brand congruity, but not with actual self-brand congruity. These moderately positive
correlations indicate that as a participant perceives increased self-brand congruity (both
actual and ideal), his adoption behavior likelihood also increases. Starbucks’s
correlation strengths were consistently greater for actual self-brand congruity, whereas
Dunkin’ Donuts showed the opposite to be true, though only two of the Dunkin’ Donuts
correlations were significant. Overall, this test showed that both actual and ideal
self-brand congruity were positively related to adoption behavior likelihood. Hypothesis
1 was accepted as a result.
• Hypothesis 2: Brand-SNS compatibility will be positively related to the likelihood of SNS brand adoption behavior.
Just as in the analysis for self-brand congruity in Hypothesis 1, brand-Facebook
compatibility for Hypothesis 2 was also recoded such that increased compatibility was
associated with increasing numerical values (“1” = “strongly disagree” and
“5” = “strongly agree”) to keep positive correlations depicting a positive relationship.
Hypothesis 2 was explored with the same process as the latter part of Hypothesis 1 with
bivariate correlations. As with Hypothesis 1, comparisons with brand-Facebook
compatibility were observed with both intent to add the brand to one’s Facebook profile
and positive action items indicated by the participant that they would likely take, as well
57
as the calculated adoption behavior likelihood (Table 4-10). Since t-tests showed
significant differences between groups for brand-Facebook compatibility, each brand
was observed separately in addition to the combined sample.
Observations of the intent to add the brand to their Facebook activity or profile
showed a lack of significance for weakly correlated Starbucks, Dunkin’ Donuts, and
combined samples. These same results were similarly repeated showing no
significance for the sum of positive adoptive actions and also for adoption behavior
likelihood. Thus, compatibility did not prove to be correlated, positively or negatively,
with brand adoption behavior likelihood, which denoted that a relationship between
these two variables was not present. Unlike self-brand congruity results in Hypothesis 1,
brand-Facebook compatibility did not demonstrate a significant positive relationship with
adoption behavior likelihood, resulting in the rejection of Hypothesis 2.
Interaction Effect of SNS Motivations
The secondary aspect explored in this study was the interaction effect of four
usage motivations for SNSs, including entertainment, information, social interaction, and
convenience. The dependent variable for the following two hypotheses and two
research questions was the calculated overall adoption behavior likelihood addressed
previously in Hypotheses 1 and 2. Since the independent and interaction variables were
inversely coded from the dependent variable, a negative coefficient indicated a positive
relationship, and vice versa, for each of the four motivational effects explored in
Hypothesis 3, Hypothesis 4, Research Question 1, and Research Question 2.
Differences between Starbucks and Dunkin’ Donuts’s samples for the individual
entertainment, social interaction, and convenience motivations were not significant. As a
result, a greater focus was placed on the combined sample for these examinations than
58
for the information motivation. The differences between groups for the information
motivation were significant, meaning that a greater weight needed to be assigned to
each brand individually, especially for the interaction with self-brand congruity
measures. Since significant differences occurred between groups for both actual and
ideal self-brand congruity, brand groups were still given individual attention for not only
the information motivation, but also the entertainment, social interaction, and
convenience motivations.
• Hypothesis 3: An entertainment SNS motivation will positively interact with both actual and ideal self-brand congruity in predicting SNS adoption behavior likelihood.
Hypothesis 3 addressed interaction effects between an entertainment motivation
and self-brand congruity measures with a multiple regression model (Table 4-11). The
independent variables for this analysis were actual and ideal self-brand congruity,
entertainment motivation, and the interaction effect between the entertainment
motivation and both actual and ideal self-brand congruity.
The combined sample equation was statistically significant [F(5,68) = 4.06*,
*p < .01]. This model produced an R value of .48, indicating a moderate correlation
between actual and ideal self-brand congruity, entertainment motivation, both their
interaction effects, and the dependent variable, adoption behavior likelihood. R-squared
was .23, meaning that 23% of variance in these five variables is explained by adoption
behavior likelihood. Likewise, the Starbucks equation was also statistically significant
[F(5,35) = 5.12*, *p < .01] with a moderate correlation of .65 and 42% of variance
explained by the dependent variable. Since both of these equations were significant, a
linear relationship occurred between the variables for each sample group and these two
equations can be projected onto the population. In contrast to the Starbucks and
59
combined samples, the Dunkin’ Donuts equation was not statistically significant, and
thus, no linear relationship was present for this sample.
Although both the Starbucks and combined sample equations proved significant,
independently, the only significant relationship with adoption behavior likelihood was for
entertainment motivation for the Starbucks sample (t = -2.96, p < .05). However, this
means that for each sample group, both the individual self-brand congruity variables
and the interaction effects between entertainment motivation and actual and ideal
self-brand congruity were not significant. Thus, the two interactions between the
entertainment motivation and both actual and ideal congruity have no unique
contribution to predicting a Y-value for the regression equation. As a result, Hypothesis
3 is rejected.
• Hypothesis 4: An information SNS Motivation will positively interact with both actual and ideal self-brand congruity in predicting SNS adoption behavior likelihood.
The same multiple regression tests from Hypothesis 3 were applied in the
exploration of Hypothesis 4 with entertainment motivation replaced with information
motivation. The independent variables for this analysis were actual and ideal self-brand
congruity, information motivation, and the two interaction effects between information
motivation and the two self-brand congruity components (Table 4-12). Significant
differences between brands for both the individual information motivation variable and
the self-brand congruity measures provided reason to carefully consider each brand on
an individual basis.
For this hypothesis, Dunkin’ Donuts returned a regression equation that was not
significant. In addition, this examination produced no significant individual relationships
with adoption behavior likelihood for the Dunkin’ Donuts sample. However, as with the
60
previous motivation, both Starbucks [F(5,35) = 7.66*, *p <.01] and the combined sample
[F(5,68) = 6.13*, *p <.01] produced significant linear regression equations. Thus, linear
relationships are present for these sample groups and these regression equations can
be projected onto the population. Starbucks showed a strong correlation (R = .72)
between the inputted self-brand congruity, motivation, and interaction variables, and the
dependent variable, with 52% of variance explained by adoption behavior likelihood.
Likewise, the combined sample revealed a moderate correlation between these
variables (R = .56), explaining 31% of the variance present.
The individual information motivation variable was significantly related to adoption
behavior likelihood for both the combined sample (t = -3.12, p < .01) and the Starbucks
sample (t = -3.04, p < .01). In fact, information motivation was the only variable with a
significant relationship with the dependent variable for the combined sample. The
negative coefficients for the information variable for these sample groups indicate a
positive effect on adoption behavior likelihood.
Furthermore, while Dunkin’ Donuts produced no significant individual relationships,
Starbucks returned actual self-brand congruity (t = -2.87, p < .05) and the interaction
variable between actual congruity and information motivation (t = 2.18, p < .05) to also
have significant relationships with adoption behavior likelihood. Although not significant
for the combined sample, these variables were the most important variables for the
model with Beta values of -1.43 and 1.35, respectively. Partially consistent with
predictions for these variables, the negative coefficient for actual congruity implied that
actual self-brand congruity is positively related to adoption behavior likelihood.
61
However, the positive coefficient for the interaction between actual congruity and
information motivation indicated a negative effect.
The other half of the self-brand congruity measures (ideal congruity) showed no
significant correlations with adoption behavior likelihood individually nor by interacting
with information motivation. Thus, these variables had no unique contribution to
predicting a Y-value for the model’s regression equation. As a result, a positive
information motivation interaction with both actual and ideal self-brand congruity cannot
be statistically confirmed and Hypothesis 4 is rejected. Supplementary to this, the
differences between Starbucks and Dunkin’ Donuts’ results signify that the effect of
information motivation and the interaction effect between information motivation and
self-brand congruity vary per brand, so each brand needs to be considered on an
individual basis in future studies.
Research Questions
The research questions that naturally follow these hypotheses are simply a
continuation of the interaction effects of the final two SNS usage motivations, social
interaction, and convenience. Since these variables were expected to have negative
impacts on behavior toward a brand, they were posed as research questions to be
explored, not to provide definitive statistical answers. However, multiple regression
analyses were performed for both the social interaction motivation and convenience
motivation in the same manner as in Hypothesis 3 and Hypothesis 4, but with expected
negative influences rather than positive influences.
• Research Question 1: Do social interaction SNS motivations negatively interact with actual and ideal self-brand congruity in predicting SNS brand adoption behavior likelihood?
62
Research Question 1 entered actual and ideal self-brand congruity, social
interaction motivation, and the two interactions between each self-brand congruity
component and social interaction motivation to have potential significant relationships
with adoption behavior likelihood (Table 4-13). Social interaction motivations were
expected to negatively impact the effect of both actual and ideal self-brand congruity on
adoption behavior likelihood. Significant differences between brands for the self-brand
congruity measures provided reason to carefully consider each brand on an individual
basis.
As with the combined samples of Hypothesis 3 and Hypothesis 4, the combined
sample returned a significant linear regression equation [F(5,68) = 5.28*, *p <.01], but
no individual significant individual relationships, meaning that they had no unique
contribution to predicting a Y-value for the regression equation. However, Starbucks
and Dunkin’ Donuts showed varying results.
Starbucks returned both a significant regression equation [F(5,35) = 9.29*,
*p < .01], as well as individual significant correlations between independent variables
and adoption behavior likelihood. The regression equation indicated a strong correlation
between the independent and dependent variables (R = .76), accounting for 57% of the
variance present. The significant relationships with adoption behavior likelihood for this
sample were with actual self-brand congruity (t = -3.39, p < .01), the independent social
interaction motivation (t = -3.66, p < .01), and the interaction between social interaction
motivation and actual congruity (t = 2.82, p < .01). Consistent with expectations, the
positive coefficient for the interaction between ideal self-brand congruity and social
interaction motivation indicates a negative influence on adoption behavior likelihood.
63
Dunkin’ Donuts also returned significant individual variables, including both actual
congruity (t = 2.11, p < .05), ideal congruity (t = -2.05, p < .05), and the interaction effect
between social interaction motivation and actual congruity (t = -2.01, p < .05). The
negative coefficient for the interaction between actual self-brand congruity and social
interaction motivation reveals a positive influence on adoption behavior likelihood, which
contradicts the predicted negative impact, and contrasts the Starbucks result. However,
the overall regression equation for the Dunkin’ Donuts sample was not significant. Thus
no linear relationship existed between the inputted variables and adoption behavior
likelihood, and the equation cannot be projected onto the population.
While the Starbucks and combined sample models were statistically significant,
individual variables complicated this analysis. For both brands, social interaction did
significantly interacted with actual self-brand congruity. However, besides the fact that
the directional influences of these two results contradicted each other, the second
self-brand congruity component, ideal congruity, revealed that it did not have a
significant interaction effect with social interaction motivation for either brand.
Consequently, the data do not confirm that social interaction motivation negatively
interacts with both self-brand congruity components in predicting adoption behavior
likelihood. However, the significant interaction effects between actual self-brand
congruity and social interaction motivation give reason to explore this avenue further.
Actual self-brand congruity may be a significant factor in this model, whether it is
independently or interacting with social interaction motivation.
• Research Question 2: Do convenience SNS motivations negatively interact with actual and ideal self-brand congruity in predicting SNS brand adoption behavior likelihood?
64
The final motivation observed was convenience motivation and its interaction
effect with actual and ideal self-brand congruity on adoption behavior likeliness (Table
4-14). Variable effects were observed individually per brand due to significant
differences observed between groups for self-brand congruity measures.
As with the previous motivations, the equations for both Starbucks
[F(5,35) = 7.35*, *p < .01] and the combined sample [F(5,68) = 5.45*, *p < .01] were
statistically significant allowing the linear relationship to be projected onto the
population. The regression equation for Dunkin’ Donuts was not significant and
therefore cannot be projected onto the population. Starbucks showed a strong
correlation (R = .72) accounting for 51% of variance. The combined sample followed
with a moderate correlation (R = .54) with 29% of variance accounted for by adoption
behavior likelihood.
Although Dunkin’ Donuts did not return any of the independent or interaction
variables to be significant, Starbucks and the combined sample showed discrepancies
on this matter. Similar to the information motivation results, both Starbucks (t = -3.97,
p < .01) and the combined sample (t = -2.81, p < .05) returned the independent
convenience motivation as significantly correlated with adoption behavior likelihood.
This was the only significant relationship for the combined sample, but the data for the
Starbucks sample illustrated that in addition to the independent convenience motivation,
its interaction effect with ideal self-brand congruity also had a significant impact on
adoption behavior likelihood (t = 2.04, p < .05).
Negative coefficients for both the combined sample and Starbucks independent
convenience motivation variable implied a positive effect on adoption behavior
65
likelihood, contrary to expectations. On the other hand, the interaction effect between
convenience motivation and ideal congruity for the Starbucks sample displayed a
positive coefficient, meaning that convenience motivation negatively interacted with
ideal self-brand congruity in predicting adoption behavior likelihood.
Despite the fact that this component of the interaction variable set was consistent
with predictions for this research question, it was the only interaction effect for any of
the sample groups that was significant. The lack of significance for this motivation’s
interactions with actual congruity for Starbucks, and with both actual and ideal congruity
for Dunkin’ Donuts and the combined sample, indicate that they have no unique
contribution to predicting a Y-value in the regression model equation. Hence, this
multiple regression analysis revealed that while the convenience motivation may
individually have a significant relationship with adoption behavior likelihood, no
significant negative interaction effect between convenience motivation and self-brand
congruity components on adoption behavior likelihood was confirmed. This was true of
all four motivational interaction variables examined in this study. Also, similar to the
previous three motivations examined, the differences between Starbucks and Dunkin’
Donuts’s results indicated that the effect of convenience motivation and its interaction
effect with self-brand congruity measures varied according to brand.
A summary of the significant relationships with adoption behavior likelihood for
each motivation’s multiple regression analysis made in Hypothesis 3, Hypothesis 4,
Research Question 1, and Research Question 2 for the combined sample (labeled
as”‘C.S.”), Starbucks (labeled as “S”), and Dunkin’ Donuts (labeled as “D.D.”) indicates
that social interaction had the highest number of significant relationships, followed by
66
information, convenience, and lastly, entertainment (Table 4-15). A look at each sample
indicates that Starbucks returned the most variables significantly related to adoption
behavior likelihood, including all four independent motivation variables. Actual
self-brand congruity proved to have a significant relationship with adoption behavior
likelihood more often than ideal self-brand congruity. Interaction effects showed this
same trend.
Limitations
This study is limited by several factors. First, relatively low variances accounted for
by the models act as a noteworthy hindrance in this study, as do the low reliabilities of
the individual motivation and compatibility variables. Second, the sample includes
students from the College of Journalism and Communication at the University of Florida.
These students, being already clearly interested in communication, may be more
interested and involved with SNSs and higher posting behavior than students in other
disciplines. Third, the issue of online privacy could have been an influential factor, as
indicated by many responses about current profile habits, but was too broad to explore
in the scope of this study with due diligence.
Fourth, due to the nature of the study where qualifying components had to be met
to fulfill the questionnaire to completion, the resulting sample size was lower than
preferred which lead to questionable reliability measures. Fifth, for the portion of the
sample that was extended on Facebook, a more reliable sample would have been
beneficial from a more national and global audience. If this were the case, brands
chosen for observation would need to be represented in those selected geographical
areas. Sixth, due to limiting the length of the questionnaire, brands were chosen to be
restricted to one product category and participants were presented with only one of the
67
two brands. A broader brand collection and a direct comparison between brands
presented at once would have been helpful in achieving a broader perspective of
branded behavior.
The questionnaire itself included several limitations. In addition to the items
mentioned previously, if it had been possible to return to the questionnaire and modify it,
changes would have included discarding the coffee drinker qualification to increase
sample size, since many non-coffee drinkers also have developed attitudes toward
well-known brands and their typical users. A modified questionnaire also would have
included more questions to measure compatibility to obtain a reliability score for the
variable. Finally, the question regarding posting behavior was limited by an unequal
number of items for positive, negative, and neutral responses. As a result, only positive
actions were sufficiently observed. Future research observations regarding simple
action versus inaction and positive action versus negative action could provide added
valuable insight on this subject. Such research modifications would likely introduce new
variables in the effect of brand on self-brand congruity measures and adoptive SNS
posting behavior that were beyond the scope and capabilities of this study.
68
Table 4-1. Sample profile summary statistics Starbucks
(n=41) Dunkin’ Donuts (n=33)
Total sample (n=74)
Sig.
Mean S. D. Mean S. D. Mean S.D. Age 22.15 3.69 23.12 4.21 22.58 3.93 NS* Age group # % # % # % NS** 18-24 35 85.4 25 75.8 60 81.1 25-34 5 12.2 7 21.2 12 16.2 35+ 1 2.4 1 3.0 2 2.7 Gender # % # % # % NS** Male 15 36.6 11 33.3 26 35.1 Female 26 63.4 22 66.7 48 64.9 Education # % # % # % NS** High School 10 24.4 2 6.1 12 16.2 AA in progress 1 2.4 -- -- 1 1.4 AA -- -- 3 9.1 3 4.1 Bachelor in progress 22 53.7 17 51.5 39 52.7 Bachelor 5 12.2 5 15.2 10 13.5 Graduate 2 4.9 6 18.2 8 10.8 PhD in progress 1 2.4 -- -- 1 1.4 Ethnicity # % # % # % NS** Asian 1 2.4 2 6.1 3 4.1 Black/African American 4 9.8 4 12.1 8 10.8 Hispanic/Latino 9 22.0 3 9.1 12 16.2 White/Caucasian 26 63.4 24 72.7 50 67.6 Other 1 2.4 -- -- 1 1.4
69
Table 4-1. Continued Starbucks
(n=41) Dunkin’ Donuts (n=33)
Total sample (n=74)
Sig.
Mean S. D. Mean S. D. Mean S.D. Facebook familiaritya 1.51 .60 1.48 .83 1.50 .70 NS* Facebook acceptabilitya 1.76 .70 194 .79 1.84 .74 NS* Facebook likely to visita 1.61 .89 1.30 .47 1.47 .74 NS* Facebook frequency # % # % # % NS** Multiple times/day 29 70.7 25 75.8 54 73.0 Once per day 4 9.8 5 15.2 9 12.2 Multiple times/week 4 9.8 3 9.1 7 9.5 Once per week 2 4.9 0 0.0 2 2.7 One per month 1 2.4 0 0.0 1 1.4 LT once per month. 1 2.4 0 0.0 1 1.4 Facebook current behavior # % # % # % NS** Browsing 15 36.6 10 30.3 25 33.8 Interacting 22 53.7 17 51.5 39 52.7 Adding profile info 4 9.8 6 18.2 10 13.5 Mean S. D. Mean S. D. Mean S.D. Facebook total activity timeb 6.69 10.58 5.35 4.43 6.09 8.39 NS* *One-way ANOVA **Chi-square test aFacebook familiarity. Facebook acceptability and Facebook likely to visit were measured on scale of 1 to 5 where “1” = very familiar, very acceptable, and definitely will visit bFacebook total activity time is reported in hours, e.g, 6.69 = 6 hours and .69 of an hour (41.4 minutes)
70
Table 4-2. Current Facebook profile behavior (direct expression) Starbucks Dunkin’ Donuts Total sample Sig. Facebook profile- basic # % # % # % NS Full 8 19.5 11 33.3 19 25.7 Almost full 15 36.6 12 36.4 27 36.5 Very limited 16 39.0 10 30.3 26 35.1 None 2 4.9 -- -- 2 2.7 Facebook profile- personal # % # % # % NS Full 2 4.9 4 12.1 6 8.1 Almost full 12 29.3 7 21.2 19 25.7 Very limited 23 56.1 21 63.6 44 59.5 None 4 9.8 1 3.0 5 6.8 Facebook profile- contact info # % # % # % NS Full -- -- 1 3.0 1 1.4 Almost full 8 19.5 7 21.2 15 20.3 Very limited 27 65.9 21 63.6 48 64.9 None 6 14.6 4 12.1 10 13.5 Facebook profile- education/work # % # % # % NS Full 8 19.5 8 24.2 16 21.6 Almost full 13 31.7 11 33.3 24 32.4 Very limited 18 43.9 12 36.4 30 40.5 None 2 4.9 2 6.1 4 5.4
71
Table 4-3. Current Facebook profile behavior (indirect expression) Starbucks Dunkin’ Donuts Total sample Sig. Facebook profile- groups # % # % # % NS Many 15 36.6 10 30.3 25 33.8 A few 23 56.1 20 60.6 43 58.1 None 3 7.3 3 9.1 6 8.1 Facebook profile- fan items # % # % # % NS Many 4 9.8 5 15.2 9 12.2 A few 18 43.9 17 51.5 35 47.3 None 18 43.9 11 33.3 29 39.2 Not familiar 1 2.4 -- -- 1 1.4 Facebook profile- bumpers from others # % # % # % NS Many 7 17.1 7 21.2 14 18.9 A few 18 43.9 8 24.2 26 35.1 None 15 36.6 18 54.5 33 44.6 Not familiar 1 2.4 -- -- 1 1.4 Facebook profile- bumpers by self # % # % # % NS Many 2 4.9 3 9.1 5 6.8 A few 10 24.4 3 9.1 13 17.6 None 28 68.3 27 81.8 55 74.3 Not familiar 1 2.4 -- --1 1.4 Facebook profile- gifts from others # % # % # % NS Many 1 2.4 6 18.2 7 9.5 A few 29 70.7 9 27.3 38 51.4 None 11 26.8 18 54.5 29 39.2 Facebook profile- other # % # % # % NS Many 2 4.9 3 9.1 5 6.8 A few 26 63.4 18 54.5 44 59.5 None 13 31.7 12 36.4 25 33.8 Mean S. D. Mean S. D. Mean S.D. Sig. Total current Facebook activitya 23.83 4.33 23.30 3.95 23.60 4.14 NS
aThe sum of all Facebook actions from Tables 4-2 and 4-3 where max = 44 and min = 10
72
Table 4-4. Brand attitudes, brand congruity, and compatibility
Starbucks (n=41)
Dunkin’ Donuts (n=33)
Total sample (n=74)
Mean S. D. Mean S. D. Mean S.D. Sig. Brand familiarity 1.76 .97 2.09 .91 1.91 .95 NS Brand attitude indexa 1.98 .59 2.35 .80 2.15 .71 p<.05 Good-bad 2.00 .71 2.33 .96 2.15 .84 -- Pleasant-unpleasant 1.95 .74 2.39 .70 2.15 .75 -- Favorable-unfavorable 2.20 .78 2.52 .94 2.34 .86 -- Likeable-unlikeable 2.00 .71 2.42 .97 2.19 .85 -- Alpha .92 .97 .95 -- Actual self-brand congruityb 2.56 1.05 3.18 1.16 2.84 1.14 p<.05*
Ideal self-brand congruityc 2.76 1.11 3.30 1.21 3.00 1.18 p<.05**
Pearson correlation .83 .81 .83 -- Compatibilityd 2.46 .67 2.82 .95 2.62 .82 p<.05
aAttitude toward Starbucks was significantly more positive than attitude toward Dunkin’ Donuts (t = -2.29; df = 72; p < .05) bStarbucks actual self rating was significantly greater than Starbucks ideal self rating (t = -1.95, df = 40; p < .05) (paired t-test) cDunkin’ Donuts actual self rating was not significantly different from its ideal self rating (paired t-test) *Actual self-congruity was significantly greater for Starbucks than Dunkin’ Donuts (t = 2.39, df = 72; p < .05). **Ideal self-congruity was significantly greater for Starbucks than Dunkin Donuts (t = 2.00, df = 72; p < .05). dCompatibility between Starbucks and Facebook was significantly higher than compatibility between Dunkin’ Donuts and Facebook (t = -1.88; df = 72; p < .05)
73
Table 4-5. Summary statistics for independent and motivation variables Starbucks Dunkin’ Donuts Total sample
Sig. Independent variables Mean Reliability Mean Reliability Mean Reliability Brand attitude 1.98 α = .92 2.35 α = .97 2.15 α = .95 p < .05a Actual congruity 2.56
R = .83 3.18
R = .81 2.84
R = .83 p < .05b
Ideal congruity 2.76 3.30 3.00 p < .05c
Compatibility 2.46 NA1 2.82 NA1 2.62 NA1 p < .05d Motivation variables Mean Reliability Mean Reliability Mean Reliability Sig. Entertainment 1.90 α = .72 1.89 α = .56 1.90 α = .65 NS Convenience 2.17 α = .64 2.20 α = .68 2.19 α = .65 NS Social interaction 2.59 α = .53 2.46 α = .48 2.54 α = .50 NS Information 3.63 α = .86 3.27 α = .70 3.47 α = .82 p < .05e aAttitude toward Starbucks was significantly more positive than attitude toward Dunkin’ Donuts (t = -2.29; df = 72; p < .05) b Actual self-congruity was significantly greater for Starbucks than Dunkin’ Donuts (t = 2.39, df = 72; p < .05). c Ideal self-congruity was significantly greater for Starbucks than Dunkin Donuts (t = 2.00, df = 72; p < .05). dCompatibility was measured by a single 1-5 item where 1 = Very Compatible. Compatibility between Starbucks and Facebook was significantly higher than compatibility between Dunkin’ Donuts and Facebook (t = -1.88; df = 72; p < .05) eInformation motivation was significantly higher for Dunkin’ Donuts participants than for Starbucks participants (t = 1.74; df = 72; p < .05) Table 4-6. Dependent variable component correlations
Dependent variable Starbucks Dunkin’ Donuts Total sample R Sig. R Sig. R Sig.
Intent to add brand to Facebook with adoptive behavior
.71 p < .05 .72 p < .05 .72 p < .05
Brand attitude with combined adoption behavior likelihood
.71 p < .05 .38 p < .05 .56 p < .05
74
Table 4-7. Dependent variable measures
Starbucks Dunkin’ Donuts Total sample Mean Alpha Mean Alpha Mean Alpha Sig.
Likely of adding (single semantic scale: 5 = “Definitely”)
2.20 NA 1.82 NA 2.03 NA NS
Positive adoptive behavior (n=10) # % # % # % Sig. Accept and display a brand logo gift that someone sent you
16 39.0 8 24.2 24 32.4 NS
Write/post/display something positive about brand in profile or status
13 31.7 7 21.2 20 27.0 NS
Become a fan of the brand on Facebook 9 22.0 9 22.0 18 24.3 NS Send a brand logo gift to someone 10 24.4 3 9.1 13 17.6 p<.05 Add/display Facebook bumper sticker of brand logo someone sent you
10 24.4 3 9.1 13 17.6 p<.05
Accept an invitation to join a group with positive attitude toward brand
8 19.5 4 12.1 12 16.2 NS
Send someone else a bumper sticker of the brand logo
7 17.1 2 6.1 9 12.2 NS
Add/display Facebook bumper sticker of brand logo you found
5 12.2 1 3.0 6 8.1 NS
Search out brand Facebook profile and more info
4 9.8 1 3.0 5 6.8 NS
Search out and join group with positive attitude toward the brand
2 4.9 1 3.0 3 4.1 NS
Mean S.D. Mean S.D. Mean S.D. Sig.
Total positive behaviors 2.05 2.29 1.18 1.69 1.66 2.08 p<.05* *Total positive behaviors regarding Starbucks were significantly higher than total positive behaviors regarding Dunkin’ Donuts (t = 1.87; df = 72; p < .05)
75
Table 4-8. Correlations Self-brand
congruity Current behavior
Brand attitude
Adoption behavior likelihood1 Actual Ideal
Combined sample
Actual congruity 1.00 .83** .07 .58** -.43** Ideal congruity .83** 1.00 .12 .57** -.39** Current behavior .07 .12 1.00 .13 -.27* Brand attitude .58** .57** .13 1.00 -.56** Adoption behavior likelihood1 -.43** -.39** -.27* -.56** 1.00
Starbucks Actual congruity 1.00 .83** -.03 .60** -.47** Ideal congruity .83** 1.00 .06 .48** -.36* Current behavior -.03 .06 1.00 .17 -.28 Brand attitude .60** .48** .17 1.00 -.71** Adoption behavior likelihood -.47** -.36* -.28 -.71** 1.00
Dunkin’ Donuts
Actual congruity 1.00 .81** .25 .50** -.29 Ideal congruity .81** 1.00 .24 .60** -.37* Current behavior .25 .24 1.00 .14 -.32 Brand attitude .50** .60** .14 1.00 -.38* Adoption behavior likelihood -.29 -.37* -.32 -.38* 1.00
1Negative correlation values indicate positive relationships due to inverse coding **. Correlation is significant at the 0.01 level (2-tailed) *. Correlation is significant at the 0.05 level (2-tailed) Table 4-9. Self-brand congruity correlations Hypothesis 1 Starbucks Dunkin’ Donuts Total sample Actual self-brand congruity and intent to add brand to Facebook .49** .33 .44**
Ideal self-brand congruity and intent to add brand to Facebook .39* .37* .41**
Actual self-brand congruity and positive action items .42** .24 .38**
Ideal self-brand congruity and positive action items .31* .33 .35**
Actual self-brand congruity and adoption behavior likelihood .47** .29 .43**
Ideal self-brand congruity and adoption behavior likelihood .36* .37* .39**
**. Correlation is significant at the 0.01 level (2-tailed) *. Correlation is significant at the 0.05 level (2-tailed)
76
Table 4-10. Compatibility correlations
Hypothesis 2 Starbucks Dunkin’ Donuts Total sample
R Sig. R Sig. R Sig. Compatibility and intent to add brand to Facebook
.26 .10 .20 .27 .07 .57
Compatibility and positive action items
.21 .19 .28 .12 .04 .76
Combined intention and adoption behavior likelihood
.24 .13 .26 .14 .05 .68
77
Table 4-11. Multiple regression for entertainment motivation variable (H3)
78
Table 4-12. Multiple regression for information motivation variable (H4)
79
Table 4-13. Multiple regression for social interaction motivation variable (RQ1)
80
Table 4-14. Multiple regression for convenience motivation variable (RQ2)
81
Table 4-15. Summary of multiple regression significant relationships
Motivations
Self-brand congruity Individual motivation
Interaction effect Actual Ideal Actual Ideal C.S. S D.D. C.S. S D.D. C.S. S D.D. C.S. S D.D. C.S. S D.D.
Entertainment X Social interaction X X X X X X Information X X X X Convenience X X X
82
CHAPTER 5 DISCUSSION AND CONCLUSIONS
Discussion
The main purpose of this research was to determine if an opportunity occurs for
greater customer relationship management and to clarify a few of the fundamental
principles that can lead to successful relationships with consumers via SNSs. To do so,
this study examined the relationship of actual and ideal self-brand congruity, as well as
brand-SNS compatibility with consumer likelihood of publicly associating themselves
with a brand on Facebook. In conjunction with these variables, motivational interactions
with each self-brand congruity measure were examined to attempt to further explain
what impacts SNS posting behavior. The two brands utilized to exemplify these
potential relationships were Starbucks and Dunkin’ Donuts within the SNS context of
Facebook.
Two theories were applied in shaping this research. In previous research,
self-congruity theory has been used to explore brand attitude. Since brand attitude is
often used in predicting branded behavior, the linkages between these three constructs
led to an investigation in the present study of self-brand congruity’s relationships with
branded behavior within the SNS environment. This investigation was fulfilled by
observing both actual and ideal self-brand congruity perceived by participants and then
determining the presence or absence of relationships with brand adoption behavior
likelihood on Facebook. In the process, brand attitude was also measured as a source
of extra security. This examination presented results in alignment with previous
research findings, demonstrating that with an increase in brand attitude, both actual and
ideal self-brand congruity also increase.
83
Second, the uses and gratifications theory inspired an inspection of the interaction
effects of SNS usage motivations on posting behavior. Interaction effects were studied
for four motivation categories: entertainment, information, social interaction, and
convenience. These components were employed to observe each motivation’s
interaction with each self-brand congruity element in predicting adoption behavior
likelihood.
These theoretical foundations provided structure for advancing previously
explained relationships in a complex and relatively young social media environment.
The following theoretical implications detail the conclusions derived from the statistical
tests completed in Chapter 4. Collectively, they help clarify which factors contributed to
brand adoption behavior likelihood, as well as give reasons for elimination of those that
do not contribute to this behavior, as originally expected. A discussion of the
implications of these findings for marketers and advertisers in the future follows to
provide guidance for the next steps to be taken regarding this topic.
Implications
This research was fueled by an inquiry into which factors contributed to SNS
posting behavior, which indicates personal adoption of a brand. The primary constructs
observed in the present research were self-brand congruity, brand-Facebook
compatibility, and motivations for using SNS. Both self-brand congruity (actual and
ideal) and brand-Facebook compatibility were examined in Hypothesis 1 and
Hypothesis 2, respectively, using bivariate correlation analyses with regard to adoption
behavior likelihood. The four motivational interaction effects were each assigned their
own examination. Hypotheses 3 and 4 targeted the entertainment and information
motivations, respectively. Research Questions 1 and 2 addressed the predicted
84
negative interaction effects of the social interaction and convenience motivations,
respectively. To achieve the purpose of this research, a combination of frequency
analyses, descriptive summaries, t-tests, bivariate correlations, and multiple regression
analyses were implemented. Statistical compilations of these tests illustrate existent and
non-existent linear relationships between the variables.
A precautionary assessment of each sample group was performed to make sure
the brand was not overlooked as an extraneous or spurious variable. Variables were
evaluated for significant differences between the Starbucks sample and the Dunkin’
Donuts sample. These tests showed no significant differences for any of the
demographic variables or variables addressing Facebook familiarity, attitudes, and
activity. In addition, familiarity with Starbucks was not significantly different from
familiarity with Dunkin’ Donuts, which both averaged between “familiar” and “very
familiar” in responses. Thus, both Starbucks and Dunkin’ Donuts participants mirrored
each other in this sample demographically, but also in that both groups can be assumed
to be equally familiar with both Facebook and their respective brand. This established
familiarity improved the reliability of the subsequent data. Moreover, the lack of
significant differences for these variables indicated that Facebook activity and attitudes
were not contingent on varying brands. Current Facebook behavior indicated that
participants demonstrated moderate to high levels of expression behavior overall.
However, attitudes toward Starbucks were significantly more positive than those
toward Dunkin’ Donuts. Previous research establishing self-brand congruity as a
predictor of brand attitudes was further validated by this study’s high correlations
between these constructs. Both actual and ideal self-brand congruity were significantly
85
greater for Starbucks than for Dunkin’ Donuts. In this case, the results illustrate that
Starbucks participants perceived more congruity between both their actual and ideal
selves and typical Starbucks users than Dunkin’ Donuts participants perceived between
both their actual and ideal selves and the typical Dunkin’ Donuts user. Within this
Starbucks significantly greater sample, participants perceived significantly higher
congruency between their actual selves and the typical user of Starbucks than between
the typical user of Starbucks and how they would like to see themselves (ideal
congruity). However, no similar significant differences occurred between actual and
ideal self-brand congruity for the Dunkin’ Donuts sample.
Significant differences between groups signified, as with attitudes toward different
brands, that individual brands have unique levels of self-brand congruity, which applies
to both actual and ideal congruity. This finding suggests that self-brand congruity
measures are more reliant on brand attitudes, and perhaps brand personality, than
originally considered. In turn, brands that manage to achieve a high level of actual and
ideal self-brand congruity with their current or target consumers, and also achieve highly
positive brand attitudes among these consumers, could observe increased adoptive
behavior and SNS word-of-mouth benefits over brands that fail to achieve these goals.
This implication is supported further by the results of Hypothesis 1 in the following
section. The implication that follows is a need to investigate self-brand congruity
components on a per brand basis in future research. Also, based on the findings of this
research, marketers should attempt to relate their typical user image with the actual and
ideal self-images of their current and target consumers for increased word-of-mouth
benefits from SNS activity.
86
Similar to the outcomes of self-brand congruity, compatibility also varies between
brands depicted by significantly higher perceived Starbucks compatibility with Facebook
than perceived Dunkin’ Donuts compatibility with Facebook. This difference could
ultimately mean that some brands are considered more appropriate or acceptable to
talk about and display openly on an SNS than others, and that compatibility could be a
significant factor in the execution of branded Facebook behaviors.
Motivational interactions were also checked for significant differences between
brands. Among the four motivation variables, entertainment, social interaction, and
conveniences yielded no significant differences between brands. Thus, participants
were assumed to be equally likely or unlikely to identify these SNS usage motivations
for themselves regardless of the brands presented to them. Information was the only
motivation to exhibit a significantly higher information motivation for Dunkin’ Donuts than
for Starbucks. Thus, some brands elicit greater information motivation for their SNS
usage than other brands. This could mean for future research purposes that those
pursuing informational purposes are more likely to actively search for brands and carry
out associated branded behavior.
Lastly, the dependent variable also showed that adoptive behavior of Starbucks
was significantly higher than adoption of Dunkin’ Donuts into SNS behavior. Correlation
measurements showed significant strong correlations between the components used to
compute the variable. Significant moderate and strong correlations were also revealed
between the computed adoption behavior likelihood and brand attitudes. These
correlation tests substantiated the dependent variable components as good
measurements, although the relationship between brand attitude and adoption behavior
87
likelihood was significantly stronger for Starbucks than Dunkin’ Donuts. A key
observation from frequency statistics of each individual proposed action was that
participants were least likely to fulfill actions which required them to exert increased or
initiated effort in completing the behavior.
One inconsistency to note, however, was the correlations between current
Facebook behavior reported and the indicated intent of adoption behavior. As stated
previously, current Facebook behavior indicated moderate to high activity. Conversely,
adoption behavior likelihood results were comparatively much lower on their respective
scale. Although each brand individually showed no significant relationship, for the
combined sample (disregarding brand), a significant negative and weak relationship
occurred between these variables. This relationship indicated that as current Facebook
behavior increases, brand adoption behavior likelihood decreases. The result may
suggest that indicated likely behavior does not strongly translate into actual fulfilled
behavior.
The significant differences between groups for all of these studied variables are
enough to justify separate regression examinations by brand, along with the combined
sample, for each of the hypotheses and research questions. The following sections will
examine these cases in more detail.
• Hypothesis 1: Self-brand congruity (both actual and ideal) will be positively related to the likelihood of SNS brand adoptive behavior.
The goal of Hypothesis 1 was to determine if a positive relationship existed
between both actual and ideal self-brand congruity and adoption behavior likelihood.
This prediction was supported in the bivariate regression test with significant moderate
positive relationships between both actual and ideal self-brand congruity and adoption
88
behavior likelihood for both Starbucks and the combined brand sample. Dunkin’ Donuts
showed only a significant positive moderate relationship for ideal congruity and adoption
behavior likelihood. Results can therefore vary, depending on brand, but overall, this
hypothesis was accepted. Higher perceived similarity between how a Facebook user
actually views himself and how he perceives the typical user of a brand indicates an
increased likelihood that the user will positively adopt the brand to his profile behavior in
some fashion. This positive relationship also applies with perceived similarity increases
between how a Facebook user would like to view himself and how he perceives the
typical user of a brand.
These results are in alignment with previous research performed by Sirgy (1997)
and Parker (2005) comparing ideal and actual self-perceptions to brand images and
personalities. The results also imply that attitudes are a key factor in the behavior of
sharing brand associations on SNSs. The correlation results for this hypothesis analysis
signify that adoption behavior likelihood is complementary with both actual and ideal
self-brand congruity cases, but not strong enough to denote complete independence in
predicting adoption behavior likelihood. The reasons for branded posting behavior thus
extend beyond simple self-brand congruity explanations, and suggest that other factors
not studied in this research are involved in the execution of such behavior. However,
with actual and ideal self-brand congruity acting as two of the variables significantly
related to adoption behavior, brands would benefit via word-of-mouth activity by aligning
themselves with the self-image perceptions of their target consumer market.
• Hypothesis 2: Brand-SNS compatibility will be positively related to the likelihood of SNS brand adoption behavior.
89
Hypothesis 2 attempted to account for the possibility that some brands are not
considered relevant or appropriate to share on SNSs. The prediction of a positive effect
of brand-Facebook compatibility on adoption behavior likelihood was rejected. Contrary
to self-brand congruity, brand-Facebook compatibility did not have a significant
relationship with adoption behavior likelihood in this bivariate correlation model for either
brand individually or the combined sample. Although significant differences occurred of
perceived compatibility between the brands, bivariate correlation results suggest that
regardless of pre-conceived image perceptions, compatibility between a brand and
Facebook have no influence on decisions to carry out brand adoptive posting behaviors
on SNSs. Speculation suggests that SNS users show a lack of concern for the
appropriateness of their posting behavior—an area where additional research could
provide more insight. An added opportunity presents itself for further research linked
with the self-brand congruity implications of determining whether specific characteristics
or personalities of brands have more impact on these variables than the brands do as a
unit. Also, this result could have been a product of observing only one brand category.
Perhaps looking at results across brand categories would be more indicative of specific
effects on behavior between brands, as denoted in the previous limitations section.
The remainder of the statistical data analyses intended to identify interaction
effects of four motivations for using an SNS. Hypotheses 3 and 4 target the predicted
positive interaction effects of entertainment and information motivations respectively.
• Hypothesis 3: An entertainment SNS motivation will positively interact with both actual and ideal self-brand congruity in predicting SNS adoption behavior likelihood.
The multiple regression analysis for Hypothesis 3 presented the individual
entertainment motivation variable for Starbucks as the only variable significantly and
90
positively related to adoption behavior likelihood out of all three sample groups. The
more a person is motivated by entertainment in their SNS usage, the more likely they
are to adopt Starbucks into his SNS posting behavior. The Starbucks regression model
showed that there was a significant moderate relationship between the independent
variables, and the dependent variable is projectable onto the population. Entertainment
motivation individually is a complementary influence on adoption behavior likelihood, but
not strong enough to act independently. This inability to act independently could mean
that factors other than SNS usage motivations are concerned. It could also indicate that
the entertainment motivation touches on a more complex motivation or set of
motivations that could have stronger significant correlations with adoption behavior
likelihood. The possible involvement of other motivations could extend differently to
varying brands. The differences in significant relationships between sample groups
indicate that some brands experience higher adoptive behavior than others when a
person is motivated to use an SNS for entertainment purposes. This divergence could
be due to differences in brand personalities and entertainment qualities attributed to a
brand.
Furthermore, neither actual self-brand congruity nor ideal self-brand congruity
were significantly related to adoption behavior likelihood. Significant relationships were
also absent among the interaction effects between both these self-brand congruity
variables and entertainment motivation. Thus, neither a positive nor a negative
interaction occurred with self-brand congruity variables in predicting adoption behavior
likelihood by entertainment motivations. It is therefore reasonable to conclude that an
entertainment motivation has no impact via interaction with self-brand congruity on the
91
likelihood of adopting the brand in SNS posting behavior. This hypothesis was rejected
as a result. The analysis results indicated that it made no difference if a person felt any
kind of congruity with a brand, for a person to adopt Starbucks into his SNS posting
behavior for entertainment purposes.
• Hypothesis 4: An information SNS motivation will positively interact with both actual and ideal self-brand congruity in predicting SNS adoption behavior likelihood.
Hypothesis 4 performed an identical analysis as Hypothesis 3, but for the
information motivation. The multiple regression tests for Hypothesis 4 returned some
results that confirmed expectations, and some results that contradicted expectations.
First, the regression model for Dunkin’ Donuts was not significant. Significant
relationships were not present for any of the individual variables inputted for this model
either. However, both the statistically significant Starbucks and combined sample
models disclosed the individual information motivation variable to have a significant
positive relationship with adoption behavior likelihood. Thus, the more users are
motivated to use an SNS for information purposes, the more likely they are to adopt a
brand into their SNS posting behavior.
Furthermore, in addition to the individual information motivation variable,
Starbucks also revealed actual self-brand congruity and its interaction with information
motivation as significantly related to adoption behavior likelihood. The effect of actual
congruity individually was positive, meaning that as perceived congruity between a
person’s actual self and the typical user of Starbucks increased, so did his likelihood of
adopting Starbucks into his SNS posting behavior. However, the interaction effect of
actual self-brand congruity and information motivation was negative, contrary to
expectations, indicating that if a person using an SNS for information purposes also
92
perceives high congruity between his actual selves and the typical Starbucks user, he is
less likely to adopt Starbucks into his SNS posting behavior.
The data detailed that regardless of any other factors, those respondents who are
motivated to use an SNS for information motivations are more likely to adopt a brand
into their posting behavior. However, when information motivation interacts with actual
self-brand congruity, an SNS user becomes less likely to adopt a brand into his posting
behavior. Many explanations could be offered for this directional discrepancy between
the individual and interactional effects of information motivation. For example, Starbucks
is a widely known brand. Perhaps, when amplified by a sense of congruity with
themselves, whom they already know rather well, users do not feel as strong a need to
collect information regarding a brand they feel they already know fairly well, as opposed
to a brand they are not as familiar with. This is substantiated by the significantly higher
information motivation observed for Dunkin’ Donuts than for Starbucks.
Moreover, the positive influence of information motivation individually could be
attributed to a rise in more interactive forum information-seeking, in which at least a
minimal level of information is often required of the information seeker to retrieve the
appropriate information. Subsequently, this informational response is likely from other
peers within an SNS user’s network. As a result, it could be likely that this type of
information seeking on SNSs plays more of a conversational role than traditional
unidirectional information retrieval on the Internet.
Furthermore, although the actual congruity and its interaction effect with
information motivation were significant for Starbucks, the remaining interaction effects
for each sample were not significantly related to adoption behavior likelihood. Therefore,
93
an information motivation to use an SNS does not significantly interact, positively or
negatively, with the self-brand congruity perceived by that user in relation to branded
adoptive behavior likelihood. Motivations outside of those examined in this study could
be more suited as interaction variables, or other factors could be key influencers in this
model, such as brand familiarity. The differences per brand for this examination indicate
that these measures are significant factors unique for individual brands, meaning that
outside factors could explain the reasons for such different responses to these
variables.
Next, Research Questions 1 and 2 approached the same type of interaction
effects as in Hypothesis 3 and Hypothesis 4 for the final two SNS usage motivations,
social interaction, and convenience. These motivations were posed as research
questions because they were predicted to have negative interaction effects with
self-brand congruity in predicting adoption behavior likelihood.
• Research Question 1: Do social interaction SNS motivations negatively interact with actual and ideal self-brand congruity in predicting SNS brand adoption behavior likelihood?
The regression model equation for the combined sample was significant, showing
a moderate positive correlation between the independent variables and adoption
behavior likelihood. However, this combined sample did not return any of the inputted
variables to have significant relationships with adoption behavior likelihood. However,
many significant individual variables existed for each of the brands individually.
Starbucks revealed three significant relationships with adoption behavior
likelihood: 1) the independent social interaction motivation which had a positive effect,
2) actual self-brand congruity which also had a positive effect, and, 3) as expected, a
negative interaction effect between these two variables. These results mean that
94
individually, if a user perceives high congruity between his actual self and the typical
user of Starbucks, he is more likely to positively adopt Starbucks into his SNS posting
behavior. Likewise, if a user is motivated in his SNS usage by social interaction
purposes individually, he is more likely to positively adopt Starbucks into his SNS
posting behavior. However, if a user is both motivated to use an SNS for social
interaction purposes and feels high similarity between how he actually views himself
and the typical user of Starbucks, then he is less likely to adopt Starbucks into his SNS
posting behavior.
There could be a couple reasons for the tendency presented by these results. One
explanation could be that when a user is highly motivated by social interaction
purposes, whether or not he feels he is similar with the typical user of a brand plays a
small role compared to when social interaction is not a high motivational priority.
Another rationale is that perhaps he is more concerned about social implications of
expressing his actual congruity with the brand. This study was performed during an
economic recession where consumers showed a sense of pride in finding great deals as
opposed to over-spending on trendy items. This kind of social economic consciousness
could have played an influential role in the perceptions of brands as well as the
willingness to associate with those brands.
Dunkin’ Donuts, on the other hand, did not return the individual social interaction
motivation variable as significantly related to adoption behavior likelihood. It did return
both actual self-brand congruity and the interaction of actual self-brand congruity and
social interaction motivation to have negative and positive significant relationships,
respectively, with the dependent variable though. Dunkin’ Donuts also found ideal
95
self-brand congruity to be significantly related positively with adoption behavior
likelihood. These results found that the more a person perceives similarity between their
actual selves and the typical user of Dunkin’ Donuts, the less likely he is to adopt
Dunkin’ Donuts into their SNS posting behavior. This could suggest a sense of
embarrassment regarding the perceived similarity, which indicates brand attitude and
perhaps brand personality play important roles in this relationship. In concert with this
conclusion, the more they perceive congruity between how they would like to view
themselves and the typical user of Dunkin’ Donuts, the more likely they are to adopt
Dunkin’ Donuts into their SNS posting behavior, which possibly indicates a sense of
pride, self-confidence, or other qualities that would be natural products of positive brand
attitudes. Lastly, if a person felt strong congruity between how he views his actual self
and how he views the typical user of Dunkin’ Donuts, and is also motivated in his SNS
usage by social interaction purposes, he is more likely to adopt Dunkin’ Donuts into his
SNS posting behavior. This implies that if a person is highly motivated by social
interaction, perhaps he is less concerned with the social implications of expressing his
self-brand congruity in some fashion.
The remainder of the variables, including the interaction effects of social
interaction with ideal self-brand congruity for each sample, attested to a lack of
significant relationships with adoption behavior likelihood. Therefore, social interaction
motivation failed to confirm a negative interaction with both actual and ideal self-brand
congruity. Result differences between brands indicate that some brands are more
influenced by social interaction and self-brand congruity than others, and each brand
96
must be considered individually on qualities such as brand attitude and brand
personality.
• Research Question 2: Do convenience SNS motivations negatively interact with actual and ideal self-brand congruity in predicting SNS brand adoption behavior likelihood?
As with the previous motivation examinations, the multiple regression models for
Starbucks and the combined sample executed for Research Question 2 exposed linear
relationships between the independent variables and adoption behavior likelihood and,
can therefore be projected onto the population. Once again, the model for Dunkin’
Donuts was not significant, so it cannot be projected onto the population. Within the
Starbucks and combined samples, the independent convenience motivation variable
proved to have a significant positive relationship with adoption behavior likelihood.
Accordingly, the more a person is motivated by convenience in his SNS usage, the
more likely he is to adopt a brand into his SNS posting activity. This shift from an
expected negative effect to a positive one could be explained by ease of use associated
with individual SNSs because this factor would likely increase or decrease usage
outcomes of convenience motivations. Another possible explanation could be linked to a
person’s SNS usage intensity. For example, if a person is already spending time on an
SNS, the convenience of already being logged in with a network and media capabilities
at his fingertips could positively influence the likelihood of adoption behavior.
In addition to the significant positive effect of convenience motivation, a significant
negative interaction effect on adoption behavior likelihood was observed by
convenience motivation and ideal self-brand congruity for the Starbucks sample. This
says that the more a person perceives similarities between how he would like to see
himself and the typical user of Starbucks and the more this person is motivated to use
97
an SNS for convenience purposes, the less likely he is to adopt Starbucks into his SNS
posting activity. One explanation for this could be that the person who views high
congruity between his ideal self and the typical user of Starbucks, is willing to extend
beyond simple convenience activity and is more likely to have other motivations than
convenience. This explanation would lead to a negative interaction with ideal self-brand
congruity’s indication of adoption behavior likelihood.
Unlike the interaction between ideal convenience motivation and ideal self-brand
congruity for the Starbucks sample, the remainder interaction effects were not
significant. Therefore, while the convenience motivation may individually be significantly
related to adoption behavior likelihood, convenience motivation does not significantly
interact, positively or negatively, with actual and ideal self-brand congruity perceived by
that user in predicting branded adoptive behavior likelihood. The lack of significant
interaction effects hint that motivations outside of those examined in this study were
more suited as interaction variables. Differences between results for Dunkin’ Donuts
and Starbucks indicate that individual brands are influenced by self-brand congruity and
convenience motivations uniquely and should be observed on a per brand basis for
future studies.
Implications Summary
Thus within the scope of this study, both actual and ideal self-brand congruity
proved to have significant positive relationships with adoption behavior likelihood
independently, while brand-Facebook compatibility did not. Together, these results
along with differences in findings per brand implicate a need for a more intense
examination of both brand attitude and unique brand characteristics or personalities and
their relationships with the variables in this study. Furthermore, actual self-brand
98
congruity had more impact on adoption behavior likelihood both individually and in
interaction with motivations.
None of the motivations significantly interacted with both the actual and ideal
aspects of a person’s self-brand congruity in predicting the likelihood of whether or not a
user adopts a brand into his SNS activity. However, all four motivations showed promise
as individual variables to have a significant relationship with adoption behavior
likelihood on a per brand basis. Despite the fact that the four motivations used were
acceptable determinants of Internet usage and SNS usage as indicated by this study,
the results shown here also imply that these SNSs motivation categories leave room for
improvement in predicting branded SNS posting behavior. Improvements may involve a
slightly modified combination of motivations that are not quite as transferrable from
usage motivations of the Internet as a whole or new motivations that have evolved from
SNS usage. However, independent of interactions with self-brand congruity, each
motivation explored in this study demonstrated at least a moderate significant
relationship with branded adoption behavior likelihood in many cases, as observed in
this study.
In alignment with the increasing popularity of SNSs and interactivity between
people and media, the implications of this excavation can serve marketers as
inspirational tools to help navigate this realm in new ways. Marketers can use this
information to improve upon relationships with their target market and hopefully to
ultimately improve upon purchase behavior by more seamlessly integrating SNS
campaigns into a marketing mix. The results of this study substantiate previous notions
about the importance of knowing the target consumer on a personal level.
99
The significant relationship between self-brand congruity and branded SNS
behavior indicates an importance for brands when forming their SNS presence to align
themselves closely with their consumers’ self-perceived qualities, both actual and ideal.
To make this alignment, first, a marketer needs to clearly define his target market. Part
of this process entails a marketer getting to know who his target market is, not just
demographically, but also on a personality and psychological level. Consumers’
entertainment, information, social interaction, and convenience tendencies should be
included in this acquaintanceship, since these motivations showed significant influences
on SNS adoption behavior unique per brand. Second, a marketer should align a brand’s
perceived typical user image with the actual and ideal perceptions and assessments of
consumers in the group he just described as his target market.
According to the results of this study, a marketer positioning a brand this way will
increase the frequency of which his brand is shared in a way that is very emotionally
attached and which has the potential to efficiently reach a very large audience. As a
result, increasing this frequency could help develop consumers of a brand simply
through exposure. Furthermore, a key element in most marketing plans is
word-of-mouth communication. In lieu of this, the mention of a brand on a user’s profile
can be especially valuable because that person is essentially attaching a trusted
endorsement not just in conversation on a singular level, but now on a platform that
reaches a more personal plural level.
Future Research
The insight obtained from this research could be just a small sample of a much
larger pool of information to be attained. Future research could include a variety of
outlooks for many questions are still left unexplained. First, this study focused only on
100
positive adoption behavior likelihood. It did not observe self-brand congruity in relation
to behavior of positive versus negative attitude or intent, which would be a natural next
step. The results of this study lend themselves to the likelihood that higher perceived
self-brand congruity results in more positive toned posting behavior, while lower
perceived self-brand congruity results in more negative toned posting behavior.
However, results opposite to this conjecture would pose a question regarding a standoff
between content attitude and simple exposure value of posting behavior as the more
beneficial component for a brand.
As the limitations section pointed out, privacy restrictions acted as a noteworthy
limitation. A Facebook user could view content only if he is friends with the individual
posting the content. This restriction limits the reach of such content. With access to
unrestricted profiles, accurate content analysis could be observed as subsequent
research. This kind of research would be uninhibited by the subjectivity of a participant
in his responses. Furthermore, privacy is a key issue to many who use SNSs and thus
may impact posting behavior that creates personal associations publicly. It could be
worth exploring the impact of such self-inflicted privacy restrictions, attitudes, and
preferences on SNS posting behavior in future research.
Also, this study could be expanded to observe relationships between the variables
in this study with purchase intent and purchase behavior. Do those who had a high
degree of actual and ideal self-brand congruity have an increased likelihood of purchase
intent? Do these people also have a higher tendency toward actual purchase behavior?
Is there a significant difference between the effect of actual congruity and ideal
congruity on purchase intent and purchase behavior? Do people with high self-brand
101
congruity indications or high adoption behavior on an SNS consume the brand with
higher frequency than those who do not post in reference to the brand or those who do
not feel a sense of self-brand congruity, actual or ideal? Does consumer loyalty depend
on whether a person posts regarding a brand first or become a consumer of the brand
first? These are just a few questions that could be explained further in subsequent
studies.
In addition, future studies could observe the interaction effects of consumers
versus non-consumers of a brand in question regarding predictive relationships
between self-brand congruity measures and adoption behavior likelihood. The current
study primarily considers consumers of a product category. However, non-consumers
are likely to know of brands they do not consume and still have brand attitudes about
them.
Finally, since only a select few of the motivation variables explored showed a
significant effect on adoption behavior likelihood, future research could attempt to clarify
more appropriate motivations for using SNSs and specifically for posting content on
SNSs. A subsequent influencing variable could be perceived genuineness. For
example, in terms of the word-of-mouth communication portrayed in branded posting
behavior, the effectiveness of this more personal form of communication could be
affected by the number of friends the poster is perceived to be delivering the message
to or the frequency with which such branded posts are made.
The research performed for this study successfully answered the questions it
posed by further clarifying factors that influenced (or did not influence) branded SNS
posting behavior. The suggested research topics would continue to provide a greater
102
knowledge of posting behavior and its value equation for marketers by taking the
implications revealed by this study and applying it to further examinations of this
subject.
103
APPENDIX A INVITATION
This is an invitation to participate in a research study. The goal of the study is to learn more about consumer’s activity on social networking sites. The information acquired will be important in expanding upon current research in advertising. Your participation in this study is greatly appreciated. It should take no more than 15 minutes and should require minimal effort, just reliable responses. Participant responses will only be used for research and will remain completely confidential and anonymous. Please pass this invitation to participate on to friends and family to complete as well. Participation is greatly appreciated.
If your first name starts with letters A-L please click link #1: http://www.surveymonkey.com/s.aspx?sm=GBfx6lx_2f6NVMiuKrayum7A_3d_3d If you first name starts with letters M-Z please click link #2: http://www.surveymonkey.com/s.aspx?sm=mBVIKb57kB6ZRTpfDzPxJA_3d_3d
(If you manually enter the link into your navigation bar, please type the URL into your browser exactly as it appears above, including the specific uppercase, lowercase, symbols, and underscores as it is shown above or else it will not work)
Please complete the survey by May 26, 2009. Surveys must be completed in their entirety in order to be valid data for this study. If you have any questions regarding the study or have trouble accessing the website links, please feel free to contact me at any time at [email protected]. Thank you for your participation!
104
APPENDIX B ONLINE SURVEY QUESTIONNAIRE
Welcome! One of the emerging methods of communication in today’s society involves social networking sites on the Internet. As a whole, these sites are an unchartered medium through which consumers and brands may connect, yet very little research has been conducted on the effects of such methods of consumer relationship development. This survey includes a wide range of questions about you and your general feelings and activity toward brands in the context of social networking sites. The questionnaire should take approximately 15-20 minutes to complete. Rest assured, you will not be asked to identify yourself individually within the survey and any information you provide will remain strictly confidential. There are no direct benefits, risks, or compensation to you for participating in the study. You may discontinue or refuse to take part at any time and your responses will not be processed unless you submit the survey upon completion. By clicking the “Submit” button, you are indicating your voluntary consent to participate in this research. Please carefully read the instructions at the beginning of each section. Most of the questions can be answered by clicking on the button(s) that best expresses your response. Questions about the study should be directed to [email protected]. Thank you very much for helping with this important survey.
Submit____ 1. Age?____
2. Do you drink coffee?
___Yes ___No
a. If Yes, what kind of coffee do you drink most often? ___(1) a purchased cup of regular coffee (caffeinated or decaffeinated roasted
coffee with or without some combination of cream and sugar) ___(2) a purchased cup of specialty coffee (includes all of the designer coffee
options such as a Latte, Frappuccino, Espresso, etcetera) ___(3) home brewed coffee
b. If Home Brewed Coffee was chosen for the previous question, what brand of
home brew do you drink most often?
___(1) Starbucks ___(2) Dunkin’ Donuts ___(3) Folgers ___(4) Maxwell House ___(5) Other (Please Specify)_____________________
105
3. Do you have an account with Facebook? If Yes, please indicate the date you joined the community in the space provided (or estimate as accurately as possible).
___(1) No ___(2) Yes Date:___________
Please answer the following questions as truthfully and accurately as you can. Clearly circle the number that you think BEST describes you. The number ‘3’ should indicate you neither agree nor disagree. “I use Facebook (because)… [coded Strongly Agree=1, Strongly Disagree=5]
4. To learn about unknown things Strongly Disagree 5 4 3 2 1 Strongly Agree
5. It’s a good way to do research Strongly Disagree 5 4 3 2 1 Strongly Agree
6. To learn about useful things Strongly Disagree 5 4 3 2 1 Strongly Agree
7. It’s convenient to use Strongly Disagree 5 4 3 2 1 Strongly Agree
8. I can get what I want for less effort Strongly Disagree 5 4 3 2 1 Strongly Agree
9. I can use it anytime, anywhere Strongly Disagree 5 4 3 2 1 Strongly Agree
10. To pass time Strongly Disagree 5 4 3 2 1 Strongly Agree
11. I just like to surf the Internet Strongly Disagree 5 4 3 2 1 Strongly Agree
12. It’s enjoyable Strongly Disagree 5 4 3 2 1 Strongly Agree
13. It’s entertaining Strongly Disagree 5 4 3 2 1 Strongly Agree
14. I wonder what other people said Strongly Disagree 5 4 3 2 1 Strongly Agree
15. To express myself freely Strongly Disagree 5 4 3 2 1 Strongly Agree
16. To meet people with my interests Strongly Disagree 5 4 3 2 1 Strongly Agree
106
17. How familiar are you with Facebook? Very Unfamiliar 5 4 3 2 1 Very Familiar
18. How acceptable is Facebook to you? Very Unacceptable 5 4 3 2 1 Very Acceptable
19. How likely are you to visit Facebook the next time you use the Internet? Very Unlikely 5 4 3 2 1 Very Likely
20. How often do you frequent Facebook?
_____(1) Multiple times per day _____(2) Once per day _____(3) Multiple times per week _____(4) Once per week _____(5) Multiple times per month _____(6) Once per month _____(7) Less than once per month
21. About how many hours and minutes would you say you actively spend on Facebook in a typical week?
Facebook Activity Time: ____hrs ____min
22. Please rank the following Facebook activities from 1 to 3. (1 meaning the activity you spend the most time on in your Facebook session, 3 being what you spend the least time doing.)
_____Adding content to your profile or other users’ profiles _____Browsing _____Responding to/interacting with others
Please indicate your level of agreement or disagreement with the following:
23. Facebook builds a relationship with me Strongly Disagree 5 4 3 2 1 Strongly Agree
24. I would like to visit Facebook again Strongly Disagree 5 4 3 2 1 Strongly Agree
25. I am satisfied with Facebook’s services Strongly Disagree 5 4 3 2 1 Strongly Agree
26. I feel comfortable in surfing Facebook Strongly Disagree 5 4 3 2 1 Strongly Agree
27. Facebook is a good place to spend my time Strongly Disagree 5 4 3 2 1 Strongly Agree
28. I would rate Facebook as one of the best Strongly Disagree 5 4 3 2 1 Strongly Agree
107
Your Facebook profile already displays: 29. Basic Information
(1)Full (2)Partial (3)Limited (4)None (5)I am not familiar with this
30. Personal Information (1)Full (2)Partial (3)Limited (4)None (5)I am not familiar with this
31. Contact Information (1)Full (2)Partial (3)Limited (4)None (5)I am not familiar with this
32. Education and Work Information (1)Full (2)Partial (3)Limited (4)None (5)I am not familiar with this
33. Groups you’ve become a member of (1)Many (2)A Few (3)None (4)I am not familiar with this
34. Items you’ve become “a fan” of (1)Many (2)A Few (3)None (4)I am not familiar with this
35. Bumper Stickers from other Facebook members (1)Many (2)A Few (3)None (4)I am not familiar with this
36. Bumper Stickers added by yourself (1)Many (2)A Few (3)None (4)I am not familiar with this
37. Gifts from other Facebook members (1)Many (2)A Few (3)None (4)I am not familiar with this
38. Other Applications (1)Many (2)A Few (3)None (4)I am not familiar with this
39. How familiar are you with [brand]?
Very Familiar 1 2 3 4 5 Very Unfamiliar
[Brand] is…
40. Extremely Good 1 2 3 4 5 Extremely Bad
41. Extremely Pleasant 1 2 3 4 5 Extremely Unpleasant
42. Extremely Favorable 1 2 3 4 5 Extremely Unfavorable
43. Extremely Likable 1 2 3 4 5 Extremely Unlikable
108
44. . 45. .
46. .
47. .
48. . 49. Would you add [Brand] to your page in any way, either visually or written verbally?
No, Definitely Not 1 2 3 4 5 Yes, Definitely
50. Consider the brand Starbucks on Facebook. Now, look through the following list and select all that apply. Consider each option on individual terms, exclusive from the others. I would: _____(-1) write/post/display something negative about [Brand] in my profile or status _____(+1) write/post/display something positive about [Brand] in my profile or status _____(0) ignore the brand and take no action _____(+1) become a fan of [Brand] _____(+1) search out [Brand] profile and more info _____(+1) add and display a bumper sticker of the [Brand] logo that you found _____(0) deny a bumper sticker of the [Brand] logo that someone else sent you _____(+1) add and display a bumper sticker of the [Brand] logo that someone else
sent you _____(+1) send someone else a bumper sticker of the [Brand] logo _____(+1) send a [Brand] logo gift to someone else _____(0) deny a [Brand] logo gift that someone else sent you _____(+1) accept and display a [Brand] logo gift that someone else sent you _____(+1) search out and join a group with a positive attitude toward [Brand] _____(-1) search out and join a group with a negative attitude toward [Brand] _____(0) ignore an invite to join a group with a positive attitude toward [Brand] _____(0) ignore an invite to join a group with a negative attitude toward [Brand] _____(-1) accept an invite to join a group with a negative attitude toward [Brand] _____(+1) accept an invite to join a group with a positive attitude toward [Brand] _____(0) Other:_____________________________________________________
[Brand]
Take a moment to think about [Brand]. Think about the kind of person who typically uses the brand. Imagine this person in your mind and then describe this person using one or two personal adjectives to describe the typical user of the brand.
44. Adjective #1 ___________________ 45. Adjective #2 ___________________
Now indicate your agreement or disagreement to the following statements:
46. The typical user of Starbucks is consistent with how I see myself. Strongly Agree 1 2 3 4 5 Strongly Disagree
47. The typical user of Starbucks is consistent with how I like to see myself.
Strongly Agree 1 2 3 4 5 Strongly Disagree 48. The Starbucks brand is compatible with Facebook.
Strongly Agree 1 2 3 4 5 Strongly Disagree
109
Demographic Questions: 51. Age _____
52. Sex
_____ (1) Male _____ (2) Female
53. What is your highest completed level of education? _____ (1) High School _____ (2) Associate Degree _____ (3) Associate Degree in progress _____ (4) Undergraduate Degree _____ (5) Undergraduate Degree in progress _____ (6) Graduate Degree _____ (7) Graduate Degree in progress _____ (8) Doctorate Degree _____ (9) Doctorate Degree in progress
54. Race/Ethnicity
_____ (1) Asian _____ (2) Black/African American _____ (3) Hispanic/Latino _____ (4) Native American _____ (5) White/Caucasian _____ (6) Other__________________
110
LIST OF REFERENCES
Aaker, D. A., Kumar, V., & Day, G. S. (2004). Marketing research (8th ed.). Hoboken, NJ: John Wiley & Sons, Inc.
About Facebook. (2004, February 4). Retrieved February 8, 2009, from www.facebook.com: http://www.facebook.com/facebook
Arndt, A. Z. (2001). New economy emotion: Engaging customer passion with e-crm. Chichester. England: John Wiley & Sons Ltd.
Arrington, M. (2009, January 22). Facebook now nearly twice the size of MySpace Worldwide. Retrieved August 18, 2009, from TechCrunch: http://www.techcrunch.com/2009/01/22/facebook-now-nearly-twice-the-size-of-myspace-worldwide/
Babbie, E. (2007). The practice of social research (11th ed.). Belmont, CA: Thomson Wadsworth.
Barker, V. (2009). Older adolescents' motivations for use of SNS: The influence of gender, group identity, and collective self-esteem. International Communication Association (pp. 1-45). Montreal, Quebec, Canada: Retrieved from Communication & Mass Media Complete database.
Bauman, Z. (2005). Liquid life. Cambridge, England. UK: Polity Press.
Beavis, C. (2009). Games within games. In R. Willett, M. Robinson, & J. Marsh (Eds.), Play, creativity and digital cultures (pp. 15-35). New York: Routledge.
Blumler, J. G., & Katz, E. (1974). The uses of mass communications: Current perspectives on gratifications research. Beverly Hills, CA: Sage Publications.
Boyd, D. (2006). G/localization: When global information and local interaction collide. Retrieved March 23, 2009, from O'Reilly Emerging Technology Conference: http://www.danah.org/papers/Etech2006.html
Boyd, D. (2008). Can social network sites enable political action? International Journal of Media & Cultural Politics , 4 (2), 241-244.
Boyd, D. M., & Ellison, N. B. (2007). Social network sites: definition, history, and scholarship. Journal of Computer-Mediated Communication , 13 (1), 210-230.
Chartrand, T., & Bargh, J. (1999). The chameleon effect: The perception-behavior link and social interaction. Journal of Personality & Social Psychology , 76 (6), 893-910.
111
Company history. (2008). Retrieved August 18, 2009, from LinkedIn: http://press.linkedin.com/history
Company Timeline. (2009). Retrieved December 2, 2009, from Facebook: http://www.facebook.com/press/product.php#/press/info.php?timeline
Consalvo, M. (2006). Console video games and global corporations: Creating a hybrid culture. New Media and Society , 8 (1), pp. 117-137.
Consortium, C. C. (2006). About C3: Convergence. Retrieved April 27, 2009, from Massachusetts Institute of Technology: http://www.convergenceculture.org/aboutc3/index.html
Curry, L. A. (2004). Affect, decision making, and adolescent risk behavior. Gainesville, FL: University of Florida.
Davies, J. (2009). Online connections, collaborations, chronicles and crossings. In R. Willett, M. Robinson, & J. Marsh (Eds.), Play, creativity and digital cultures (pp. 108-124). New York: Routledge.
Davis, J. J. (1997). Advertising research theory and practice. Upper Saddle River, NJ: Prentice Hall.
Dong, Q., Urista, M. A., & Day, K. D. (2008). Explaining why young adults use MySpace and Facebook through uses and gratification theory. Human Communication , 12 (2), 215-229.
Douglas, J. E. (1999). The congruency between the ideal network and television network's brand personality: What makes the difference? Gainesville, FL: University of Florida.
Dowdall, C. (2009). The texts of me and the texts of us. In R. Willett, M. Robinson, & J. Marsh (Eds.), Play, creativity and digital cultures (pp. 73-91). New York: Routledge.
facebook.com - Quantcast Audience Profile. (2010). Retrieved Feb 13, 2010, from Quantcast: http://www.quantcast.com/facebook.com
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Reading, MA: Addison-Wesley.
Franken, R. (1994). Human motivation (3rd ed.). Pacific Grove, CA: Brooks/Cole.
Goodman, R. D., Maggio, J., Fedde, R. A., & Lyman, W. (2008). Growing up online . [S.I.]: PBS Video http://purl.fcla.edu/UF/lib/kidsonline/.
112
Hall, A. (2009). College students' motives for using social network sites and their relationships to users' personality traits. International Communication Association (pp. 1-38). Montreal, Quebec, Canada: Retrieved from Communication & Mass Media Complete database.
Henson, C. A. (2007). An analysis of the coffee service industry in metro manila and the buying behavior of its consumers. Quezon City, The Philippines: Loyola Schools Review.
Hoffman, D. L., & Novak, T. P. (1996). Marketing in hypermedia computer-mediated environments: Conceptual foundations. Journal of Marketing , 60 (July), 50-68.
Holzner, S. (2009). Facebook marketing: Leverage social media to grow your business. Indianapolis, IN: Que Publishing.
Jenkins, H. (2006). Convergence culture: Where old and new media collide. New York: New York University Press.
Kang, K. (1999). Interrelationship of ideal, actual, and undesired self/brand congruities across high, moderate, and low vaue-expressive products by Strausbaugh's brand personality measure. Gainesville, FL: University of Florida.
Kaye, B. K., & Johnson, T. J. (2001). A web for all reasons: Uses and gratifications of internet resources for political information. Association for Education in Journalism and Mass Communication Conference. Washington DC.
Keller, K. L. (2008). Strategic brand management: Building, measuring, and managing brand equity (3rd ed.). Upper Saddle River, NJ, New Jersey: Pearson Prentice Hall.
Ko, H., Cho, C. -H., & Roberts, M. S. (2005). Internet uses and gratifications: A structual equation model of interactive advertising. Journal of Advertising , 34 (2), 57-70.
Korgaonkar, P. K., & Wolin, L. D. (1999). A multivariate analysis of web usage. Journal of Advertising Research , 39 (2), 53-68.
Lange, R., & Lampe, C. (2008). Feeding the privacy debate: An examination of Facebook. International Communication Association (pp. 1-26). Montreal, Quebec, Canada: Retrieved from Commnication & Mass Media Complete Database.
Levi-Strauss, C. (1974). The savage mind. London: Weidenfeld and Nicolson.
Li, D. (2008). Behaviors and motivations of social networking site users: A cross-gender and cross-cultural comparison. Gainesville, FL: University of Florida.
113
Lin, C. A. (1999). Online-service adoption likelihood. Journal of Advertising Research , 39 (2), 79-89.
Lin, J.-S. (2008). Antecedents and consequences of cross-media usage [electronic resource] : A study of a TV program's official website. Gainesville, FL: University of Florida.
Lopez, K. A. (2008). Student speech rights [electronic resource]: First Amendment implications for high school students on popular online social networks, MySpace and Facebook. Gainesville, FL: University of Florida.
Lury, C. (1996). Consumer culture. Cambridge, England, UK: Polity Press.
MacKenzie, S. B., Lutz, R. J., & Park, W. P. (1989). An empirical examination of the structural antecedents of attitude toward the ad in an advertising pretesting context. Journal of Marketing , 53, 48-65.
Mitchell, A. A., & Olson, J. C. (1981). Are product beliefs the only mediator of advertising effect on brand attitudes? Journal of Marketing Research , 318-332.
Nixon, H. (2003). New research literacies for contemporary research into literacy and new media? Reading Research Quarterly , 38 (3), pp. 407-413.
Papacharissi, Z., & Rubin, A. M. (2000). Predictors of Internet use. Journal of Broadcasting and Electronic Media , 44 (2), 175-196.
Park, L., Jin, B., & Jin, S.-A. (2009). Motivations, impression management, and self-disclosure in social network sites. International Communication Association (pp. 1-36). Montreal, Quebec, Canada: Retrieved from Communication & Mass Media Complete database.
Parker, B. T. (2005). This brand's for me: Brand personality and user imagery based self-congruity. Gainesville, FL: University of Florida.
Quart, A. (2003). Branded: The buying and selling of teenagers. London: Arrow Books.
Rogers, C. R. (1959). A theory of therapy, personality, and interpersonal relationships, as developed in the client-centered framework. (I. S. Koch, Ed.) Psychology: A study of a science , pp. 184-256.
Rychlak, J. F. (1981). Introduction to personality and psychotherapy. Boston: Houghton Mifflin Company.
Severin, W. J., & Tankard, J. W. (2001). Communication theories: Origins, methods, and uses in the mass media. New York: Addison Wesley Longman.
114
Shaw, M. E., & Costanzo, P. R. (1970). Theories of social psychology. New York: McGraw-Hill.
Shirky, C. (2009, May). Clay Shirky: How social media can make history | Video on TED.com. Retrieved June 16, 2009, from TED: Ideas worth spreading: http://www.ted.com/talks/clay_shirky_how_cellphones_twitter_facebook_can_make_history.html
Sirgy, M. J., Grewal, D., Mangleburg, T. F., Park, J.-o., Chon, K.-S., Claiborne, C. B., et al. (1997). Assessing the predictive validity of two methods of measuring self-image congruence. Journal of the Academy of Marketing Science , 25 (3), 229-241.
Sirgy, M. (1986). Self-congruity: Toward a theory of personality and cybernetics. New York: Praeger Publishers.
Statistics. (2009). Retrieved December 2, 2009, from Facebook: http://www.facebook.com/facebook?v=app_7146470109&ref=pf#/press/info.php?statistics
Steinbock, D. (2000). The Birth of Internet Marketing Communications. Westport, CT: Quorum Books.
Tinkham, S., & Weaver-Lariscy, R. (1993). A diagnostic approach to assessing the impact of negative politcal television commercials. Journal of Broadcasting & Electronic Media , 37 (4), 377-398.
Trusov, M., Bucklin, R. E., & Pauwels, K. (2009). Effects of word-of-mouth versus traditional marketing: Findings from an Internet social networking site. Journal of Marketing , 73, 90-102.
Van Cleemput, K. (2009). Authenticity and subcultural style in adolescents’ self-presentation on social network sites. International Communication Association (pp. 1-18). Montreal, Quebec, Canada: Retrieved from Communication & Mass Media Complete Database.
Willett, R. (2009). Consumption, Production and Online Identities. In R. Willett, M. Robinson, & J. Marsh (Eds.), Play, creativity and digital cultures (pp. 54-67). New York: Routledge.
Wu, J. (2009). Facebook politics: An exploratory study of american youth's political engagement during the 2008 presidential election. International Communication Association (pp. 1-23). Montreal, Quebec, Canada: Retrieved from Communication & Mass Media Complete Database.
115
BIOGRAPHICAL SKETCH
Stefanie Riediger was born in Ontario, Canada, where she lived for eight years
before moving to a suburb of Cleveland, Ohio. Though she is still a Canadian citizen,
her education has taken place in American schools. Stefanie completed her
undergraduate Bachelor of Science in Business Administration degree at The Ohio
State University, where she majored in business with a specialization in marketing and
minored in visual communication design. After her graduation in the spring of 2007, she
went on to earn her Master of Advertising degree from the College of Journalism and
Communications at the University of Florida in the summer of 2010. Stefanie also
completed two internships with Bizresearch, a search engine marketing firm in
Worthington, Ohio, and Saatchi & Saatchi X, the shopper marketing division of the
global Saatchi & Saatchi advertising agency in Fayetteville, Arkansas. She is currently
pursuing a career in advertising with worldwide agencies and clients.